Viterbi Summer Undergraduate Research Experience (SURE)

• Faculty/PI: Bo Jin
• Website: composites.usc.edu
• Research Overview: Established in 2018, the mission of the lab is to address problems associated with the simulation, design, and behavior of high-performance composites and structures. The scope includes the training of graduate and undergraduate students from aerospace, mechanical, and materials engineering through teaching and sponsored research projects. Students’ research experiences lead to future opportunities including industrial internships, full-time positions, and support from faculty for fellowship and further graduate program applications, as well as opportunities for peer-reviewed publications. The students will be mentored by the faculty advisor and external industrial experts. As a member of our team, you will be working closely with a group of veteran students who are involved in the field of advanced manufacturing, machine learning, and data science. Our lab prides itself on a culture of excellence and collaboration. Since 2022 May we have been recognized with 11 research honors and national/international student competition awards. Our students receive regularly both external and internal student fellowships.
• Summer Projects
  • Additive Manufacturing of Structures with High Compressive Strength
    • Project Description: We are designing structures for high compressive strength, and they are all 3D printed. The structure should reach/beyond 10k lb. compressive strength and will need to be as tall/light as possible. We will support multiple advanced 3D printers with large print-bed and bed-heating functions for shape/warping compensations, and unlimited printing materials for the project. Satisfying structures will attend the annual SAMPE AMC additive manufacturing contest from which we have won a national 3rd in the past. Great honor and a positive plus to student’s resume.
    • Student Responsibilities: With the help of faculty and veteran students, conduct meetings and the design of your structures. Design the CAD models using whichever software you are familiar with or trained by veteran students. Print the structures using the lab printers and maintain successful printing processes.
    • Preferred Majors: Aerospace & Mechanical Engineering, Astronautical Engineering, Civil & Environmental Engineering, Industrial & Systems Engineering. EE. CS.
    • Prerequisites and Preferred Skillsets: No prerequisite skills required. In-person lab participation required.
  • Manufacturing, Processing, Testing, and Machine Learning Predictions of Composite Materials Voids and Structural Strength
    • Project Description: This project focuses on employing machine learning algorithms to detect voids in aerospace composite materials, utilizing data from micro-CT imaging to predict material and structural performance for aerospace and automotive parts.
    • Student Responsibilities: Work with external industrial advisors and conduct manufacturing, processing, and testing of composite materials and structures, and work with graduate students from USC CS department to apply machine learning algorithms developed to predict material void morphology, and structural properties.
    • Preferred Majors: Aerospace & Mechanical Engineering, Astronautical Engineering, Civil & Environmental Engineering, Industrial & Systems Engineering. EE. CS.
    • Prerequisites and Preferred Skillsets: No prerequisite skills required. In-person lab participation required.

• Faculty/PI: Yong Chen
• Website: https://viterbi-web.usc.edu/~yongchen/ | https://sites.usc.edu/cam/
• Research Overview: Manufacturing has significantly contributed to improved quality and sustainability of human life. My research goal is to advance the understanding and knowledge of Additive Manufacturing (AM) and to promote its wide applications in future engineering systems. AM is a digital manufacturing process that has become a disruptive manufacturing technology, especially for medical and aerospace industries, in which small-quantity-production is dominant. It is predicted that AM will dominate most manufacturing concerns in future mass customization. To achieve the full potential of AM in cost and time savings, the development of novel AM processes, new process control methods, and a wide selection of functional materials are critical.  In addition, AM enables revolutionary new design by using complex three-dimensional shapes, heterogeneous material properties, and multi-functionality.  Systematic knowledge regarding modeling, analyzing, synthesizing, and optimizing such product designs are required to achieve desired performance.
• Summer Projects:
  • 3D printing processes and applications
    • Description: The students will investigate new 3D printing processes developed in our lab and their applications.
    • Student Responsibilities: Students will work with Ph.D. students in the lab and perform 3D printing process development and experiments.
    • Preferred Majors: Aerospace & Mechanical Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites: The students are eager to learn, willing to work with others and enjoy hands-on work.

• Faculty/PI: Mitul Luhar
• Website: https://sites.usc.edu/fsi/
• Research Overview: The fluid-structure interactions lab combines physical experiments with reduced-complexity models to address programs in aerodynamic/hydrodynamic design, flow control, soft robotics, and environmental fluid mechanics.
• Summer Projects
  • Tunable Porous Surfaces for Turbulence Control
    • Project Description: Prior theoretical studies show that porous surfaces that are anisotropic (i.e., provide different resistance to flow in different directions) can reduce fluid dynamic drag relative to smooth, solid walls.  This reduction in drag could benefit a range of aerodynamic and hydrodynamic applications. With advances in additive manufacturing, we now have the ability to design and fabricate porous surfaces with complex geometries that provide this anisotropic response.  This project aims to design and test such surfaces in laboratory water channel experiments.
    • Student Responsibilities:  The student will be responsible for designing novel porous geometries, 3D-printing them, and testing their drag reduction performance in benchtop-scale water channel experiments.
    • Preferred Majors:  Aerospace & Mechanical Engineering, Astronautical Engineering
    • Prerequisites and Preferred Skillsets: Background in physics and calculus, computer-aided design, 3D-printing, MATLAB (or similar) are all beneficial but not required.
  • Characterizing tsunami propagation and evolution via laboratory experiments
    • Project Description: Recent work has led to the development of a novel "jet array wavemaker" that can produce prescribed waves (e.g., long waves resembling tsunamis) and currents (e.g., from a river flow) in lab-scale experiments, while simultaneously accounting for differences in seawater and freshwater density. This project will use this novel flow facility to characterize the evolution of a long saltwater wave (e.g., a tsunami) as it propagates into a freshwater flow (i.e., a river).
    • Student Responsibilities: The student will set up and run the experiment, collect state-of-the-art optical measurements (using particle image velocimetry and planar laser-induced fluorescence techniques), and analyze the resulting dataset.
    • Preferred Majors: Aerospace & Mechanical Engineering, Civil & Environmental Engineering
    • Prerequisites and Preferred Skillsets: Physics and calculus background, hands-on experimental experience (including electronics. imaging) preferred but not required.

• Faculty/PI: Jason Kutch
• Website: ampl.usc.edu
• Research Overview: The Applied Movement & Pain Laboratory (AMPL) is directed by Jason J. Kutch. Work in AMPL is at the intersection of chronic pain and movement control. We are particularly interested how brain dysfunction contributes to chronic pain. Current research in AMPL is focused on developing innovative non-pharmacologic treatments for chronic pain, including non-invasive brain stimulation and virtual reality.
• Summer Projects: We are working to develop our existing virtual reality ocean surfing simulator. We are particularly interested in students with computer graphics and game development experience in either Unity or Unreal Engine.

• Faculty/PI: Eunji Chung
• Website: https://biomaterials.usc.edu
• Research Overview: The Chung research group focuses on drug delivery, nanomedicine, and regenerative engineering to generate biomaterial strategies to address the limitations of clinical solutions, with a major emphasis on rare and genetic diseases. In particular, we are interested in biomimetic nanoparticles that can be designed to deliver molecular signals to report back on or influence the behavior of diseased tissue for biomedical applications. In addition, we are harnessing our expertise in combining bioactive scaffolds with novel stem cell sources for complex regeneration of hierarchically-ordered tissues and organs. Our group is highly interdisciplinary as our research is positioned at the intersection of engineering, biology, and medicine, and we work with a variety of collaborators to translate our materials towards clinical use
• Summer Project
  • Nanomedicine for cardiovascular disease and kidney disease
    • Project Description: Drug and gene therapy delivered by nanoparticles and biomaterials-based approaches will be developed to target diseased cells and disease pathogenesis.
    • Student Responsibilities:  Students will synthesize and characterization nanoparticles, evaluate function on cells, and support other members with in vivo therapeutic efficacy studies.
    • Preferred Majors:  Biomedical Engineering
    • Prerequisites and Preferred Skillsets: wet lab in vitro cell culture, materials characterization, HPLC, etc

• Faculty/PI: Ellis Meng
• Website: biomems.usc.edu
• Research Overview: The Biomedical Microsystems Laboratory at USC focuses on developing novel micro- and nanotechnologies for biomedical applications. In particular, we are interested in the integration of multiple modalities (e.g. electrical, mechanical, and chemical) in miniaturized devices measuring no more than a few millimeters for use in fundamental scientific research, biomedical diagnostics, and therapy.
• Summer Project:
  • Summer Research Experience in Biomedical Microdevices
    • Description: The Biomedical Microsystems Lab (https://biomems.usc.edu/) directed by Prof. Ellis Meng seeks an outstanding and highly motivated postdoctoral scholar for a leading role in developing and disseminating implantable microdevices using polymer microelectromechanical systems (MEMS) technology. The Lab is a collaborative and dynamic working environment that leverages the advantages of micro- and nanotechnologies to advance scientific discovery and healthcare.  Projects will either be in the area of neural engineering/neural interfaces and/or biomedical sensor technologies.
    • Student Responsibilities: Coordinate research tasks with graduate student, postdoc, or staff mentor. Learn protocols and standard operating procedures.  Conduct experiments.  Analyze data.  Prepare presentations and reports.  Attend lab meetings and present research progress and findings.
    • Preferred Majors: Aerospace & Mechanical Engineering,Biomedical Engineering,Chemical Engineering,Electrical & Computer Engineering
    • Preferred Skillsets: Prior work experience involving hands-on tasks. This could be research but need not be.  Laboratory experiences from class work is welcome as are hands-on experiences from student clubs, competitions, or work experiences.

• Faculty/PI: Stacey Finley
• Website: csbl.usc.edu
• Research Overview: My research group primarily works in the area of mathematical oncology, where we use mathematical models to decipher the complex networks of reactions inside of cancer cells and interactions between cells. We have combined mechanistic and data-driven modeling to study these networks and predict ways to control tumor growth. We also build models relevant for other biological systems, including endothelial cells and pancreatic beta cells.
• Summer Projects
  • Computational systems biology in cancer
    • Project Description: This project will involve predicting signaling-mediated interactions between tumor and immune cells using agent-based models and calibrating the models to tumor image data to generate reliable predictive frameworks.
    • Student Responsibilities: coding; working with senior research student
    • Preferred Majors: Biomedical Engineering, Chemical Engineering
    • Prerequisites and Preferred Skillsets: coding experience is preferred; understanding of differential equations

• Faculty/PI: Dominique Duncan
• Website: https://sites.usc.edu/duncanlab/
• Research Overview: We are a group of neuroscientists, engineers, mathematicians, computer programmers, chemists, and data scientists who share a common goal of addressing challenges to neuroscience and public health via big data. We do so by working at the intersection of neuroimaging, signal processing, informatics, and machine learning. We leverage computational tools in conjunction with mechanistically oriented neuroimaging to develop analytic tools for multimodal data, build centralized archives for data and algorithms, and promote large-scale collaborative research. Currently, we focus on applications in traumatic brain injury, epilepsy, novel devices for intracranial stimulation, and COVID-19.
• Summer Projects:
  • Multimodal analysis to study post-traumatic epilepsy
    • Project Description: The Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4R) Scientific Premise is: epileptogenesis after traumatic brain injury (TBI) can be prevented with specific treatments; the identification of relevant biomarkers and performance of rigorous preclinical trials will permit the future design and performance of economically feasible full-scale clinical trials of antiepileptogenic therapies.
    • We are contributing to the EpiBioS4Rx Consortium by identifying multimodal biomarkers of epileptogenesis. We apply data science, mathematics, and machine learning to glean insight into what factors may induce the development of seizures after brain injury so we can ultimately predict which patients will develop epilepsy. Our work centers around biomarker identification in scalp and depth electrophysiology and multimodal imaging data.
    • Student Responsibilities:  computational work, image analysis, signal processing, machine learning
    • Preferred Majors:  Biomedical Engineering,Computer Science,Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Matlab, python, R, linear algebra, machine learning

• Faculty/PI: Zhenglu Li
• Website: https://sites.usc.edu/ligroup/
• Research Overview: Our group develops and applies advanced first-principles computational approaches to study excited-state phenomena of quantum materials. We use many-electron level quantum theories with high-performance computation to accurately describe interactions among electrons and between electrons and other excitations such as phonons. We aim to provide deep understandings and novel predictions of quantum phenomena in real materials and to propose new protocols to control and engineer quantum excitations.
• Summer Project
  • Computational studies of electronic and phononic properties of quantum materials
    • Project Description: This projects applies advanced first-principles methods (density functional theory, many-body perturbation theory, etc.) to study electronic, optical, and electron-phonon coupling properties of select quantum materials (e.g., oxide superconductors, two-dimensional materials, etc.). This project aims to reveal and understand novel quantum phases arising from quantum materials, providing a comprehensive analysis to different materials properties.
    • Student Responsibilities:  The student will be responsible for carrying out the calculations and analyses, reviewing literature, and summarizing results into documented reports and presentations.
    • Preferred Majors:  Chemical Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Programing skills (e.g. Python), Data analysis, Quantum mechanics

• Faculty/PI: Yu-Tsun Shao
• Website: https://sites.google.com/usc.edu/shao-emlab/
• Research Overview: We are researchers that focus on understanding the behavior of materials and devices by developing and applying novel electron microscopy techniques. We combine imaging, diffraction, spectroscopy, and machine learning approaches to probe structural, chemical, electronic properties, and topological textures at the atomic scale.
• Summer Projects 
  • Development of multi-modal electron microscopy methods for materials characterization
    • Project Description: Electron microscopes are powerful tools that enable us to look at the microscopic details of the materials, hence provide insights into the fundamental structure-property relations. This applies to all materials -- including energy materials, quantum materials, and more. Consequently, developing new imaging modes is analogous to putting on a new pair of glasses that allows us to see things that weren't possible before, including the crystal symmetry, strain, polarity, electric or magnetic fields within the materials.
    • Through this project, we will develop new imaging modes by combining hands on experimental and data mining/machine learning efforts, to explore the materials microstructures down to the nanometer scale.
    • Student Responsibilities:  Students will have hands on experience with scanning electron microscopes, read about the principles of electron scattering and interactions with the materials, as well as learn some programming for data analysis.
    • Preferred Majors:  Aerospace & Mechanical Engineering, Astronautical Engineering, Biomedical Engineering, Chemical Engineering, Computer Science, Electrical & Computer Engineering, Industrial & Systems Engineering
    • Prerequisites and Preferred Skillsets: Python programming, experience with electronic circuits for automated controls. Students taken courses on machine learning, python programming, electron microscopy are encouraged.

• Faculty/PI: Patrick Lynett
• Website: coastal.usc.edu
• Research Overview: The faculty in this group have comprehensive expertise in understanding and engineering for coastal processes. We have active research in the topical areas of extreme coastal events such as tsunamis and cyclones, fundamental fluid mechanics and turbulence, and coastal sustainability and resilience.
• Summer Projects:
  • Ocean Wave Energy Extraction
    • Project Description:   Development of theoretical, numerical, and small-scale physical models of wave energy extraction devices.  Students would be expected to design and construct a prototype wave energy extraction device, and test the prototype for efficiency.
    • Student Responsibilities: Literature review of existing energy extraction technologies; analysis of alternatives and conceptual design of prototype; construction of prototype, including control system and electricity generation
    • Preferred Majors:  Civil Engineering; Mechanical Engineering; Environmental Engineering; Electrical Engineering
    • Prerequisites and Preferred Skillsets: Fluid Mechanics, experience with Matlab
  • Simulation of Coastal Hazards
    • Project Description: Development and application of interactive and immersive coastal hazards simulators. Topics may include virtual reality development of tsunami evacuation trainers, ship simulators in the coastal ocean, and rapid simulation of engineering projects.
    • Student Responsibilities: Coding / scripting within game development platforms; creation of shaders, scenes and objects; debugging gameplay and simulations
    • Preferred Majors: Computer Science; Civil Engineering; Mechanical Engineering; Environmental Engineering
    • Prerequisites and Preferred Skillsets: Experience with game development engines such as Unity and/or Unreal Engine

• Faculty/PI: Daniel McCurry
• Website: mccurrylab.com
• Research Overview:The McCurry lab applies the tools of organic and analytical chemistry to solve environmental problems. We primarily work in the areas of wastewater reuse and drinking water treatment.
• Summer Projects:
  • Environmental engineering and chemistry to enable safe sustainable water reuse
    • Description: Our research group currently consists of five PhD students and four undergraduate researchers who all do research in the areas of water treatment and environmental chemistry. Our major research activities include identifying the precursors and formation pathways of disinfection byproducts formed during water treatment and wastewater reuse, developing new chemical technologies for oxidation of trace organic contaminants, and developing new analytical techniques for identification and quantification of chemical pollutants using mass spectrometry. Our research primarily takes place in the BHE Water Lab, and relies heavily on recently purchased analytical instrumentation, including an ion mobility QTOF mass spectrometer, a gas chromatograph/triple quadrupole mass spectrometer, and an inductively-coupled plasma mass spectrometer. These tools allow us to quantify chemicals at extremely low concentrations (e.g., parts per trillion) and to identify previously unknown compounds in water samples.
    • Student Responsibilities: Assisting graduate students with experiments on water treatment, water reuse, and quantification of trace contaminants in water. Specific responsibilities will include preparing for and setting up batch water treatment experiments, sample preparation for mass spectrometry-based identification and quantification of pollutants, and analyzing data produced by analytical instruments. Especially dedicated students may eventually have the opportunity to advance to a fully-independent project advised directly by the PI.
    • Preferred Majors: Chemical Engineering,Civil & Environmental Engineering
    • Preferred Skillsets: General chemistry lab skills; some exposure to analytical chemistry is helpful but not required; chemistry coursework helpful

• Faculty/PI: Chukwuebuka Nweke
• Website: nwekenest.com
• Research Overview: Our research focuses on solving problems at the intersection of geotechnical engineering, earthquake engineering, seismology, and geomorphology. Current efforts focus on two thrusts:
  • 1. Seismic hazard characterization using empirical methods, physics-based simulation, and data science/machine learning. This involves taking in ground motion data (recorded earthquake information ... i.e., shaking intensity and duration) from past events and use it to develop models for probabilistic forward prediction/estimates of hazard intensity. We also focus on site characterization using various kinds of data (satellite data, measured data with in-house sensors, etc.). Our overall goal is to understand how much shaking will occur at any given location given the location of hazard sources (surrounding faults).
  • 2. Biocementation of soils for mitigation of ground failures. This involves the use of bacteria or enzymes to create calcium carbonate cement in sands, essentially converting them from sand to rock! We are studying the dynamic mechanical response (capacity to handle cyclic/earthquake shaking) of the biocemented soils as they can serve to help mitigate shaking hazards such as lateral spreading, liquefaction (sand behaving like water), and other failure hazards (erosion, dust control), or establishing innovating building materials.
  • Our goal at The N.E.S.T is to improve the engineering resiliency of civil infrastructure and the supporting systems by fostering sustainable foresight and pursuing telluric (well-grounded) collaborative excellence.

• Summer Projects
  • Biocemented Sands and Microscale Investigations to its Behavior
    • Project Description: This project is focused on understanding how microscale feature changes due to cementation and other factors affect the macro scale behavior. We will use Micro CT Scanning to help in this investigation
    • Student Responsibilities:  Processing Micro CT Scan data, conducting sand biocementation, perform resonance and stress tests.
    • Preferred Majors:  Civil & Environmental Engineering
    • Prerequisites and Preferred Skillsets: Beneficial skills include familiarity with python, R, MySQL Workbench, and image processing software (GIMP, Photoshop). You will also learn to use some of these if not fully familiar.
  • Physics-Based Earthquake Simulation Assessment and Validation in Frequency Space
    • Project Description: The project is focused on validating physics-based ground motions using real data as a reference
    • Student Responsibilities: Help process simulated waveforms, compile metadata, use data science/ML/AI to help characterize latent features
    • Preferred Majors: Civil & Environmental Engineering,Computer Science
    • Prerequisites and Preferred Skillsets: Beneficial skills include familiarity with python, R, MySQL Workbench, and ML techniques
  • Quantifying Cause of Uncertainties in Horizontal-to-Vertical Spectral Ratio (HVSR) Measurements
    • Project Description: This project is focused on quantifying/characterizing the cause of uncertainties in HVSR estimates from measurements collected using temporary and permanent sensors
    • Student Responsibilities: help conduct field experiments using in-house sensors, processing collected data and compiling it into a database
    • Preferred Majors: Aerospace & Mechanical Engineering, Civil & Environmental Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Beneficial skills include familiarity with python, R, and MySQL Workbench
  • Directional (Path) Dependencies of Sedimentary Basin Site Response
    • Project Description: This project is focused on quantifying the effects of directionality (Path Bias) on site response in large or small sedimentary basins using observed data and empirical relationships.
    • Student Responsibilities: execute statistical analysis of data relative to average or site-specific models.
    • Preferred Majors: Civil & Environmental Engineering
    • Prerequisites and Preferred Skillsets: Beneficial skills include familiarity with python (Jupyter), R, and SQL queries

• Faculty/PI: Leslie Gilliard-Abdul Aziz
• Website: https://the-sustainable-lab.com
• Research Overview: Kandis Leslie Gilliard-AbdulAziz directs the Sustainable Lab, which primarily focuses on developing novel materials for sustainable catalytic processes for low-carbon chemical production. Her primary research focus is novel materials development for CO2 sequestration and utilization using an interdisciplinary toolset from chemistry, material science, chemical, and environmental engineering.
• Summer Project
  • Optimization of Novel Materials for Carbon Capture
    • Project Description: The development of sorbents is critical for capturing greenhouse gases, such as carbon dioxide, from the air or industrial streams. However, the most prevalent issue is the capability of the sorbent to sustain performance for 100s of cycles with consistent performance. This project seeks to develop novel materials with special properties to maintain carbon capture performance for several cycles.
    • Student Responsibilities:  Wet lab synthesis of nanoparticles and bulk oxide powders
    • Materials characterization using X-ray diffraction, X-ray photoelectron spectroscopy, and Transmission electron microscopy
    • Assist with characterization of carbon capture performance
    • Preferred Majors:  Aerospace & Mechanical Engineering,Astronautical Engineering,Biomedical Engineering,Chemical Engineering,Civil & Environmental Engineering,Computer Science,Electrical & Computer Engineering,Industrial & Systems Engineering
    • Prerequisites and Preferred Skillsets: Prior coursework in general chemistry or related course; Good oral and written communication skills; Works well in teams

• Faculty/PI: Robin Jia
• Website: https://robinjia.github.io/
• Research Overview: I am interested broadly in natural language processing and machine learning, with a particular focus on building NLP systems that are robust to distribution shift at test time.
• Summer Projects:
  • Assisting human readers with question answering and fact verification
    • Description: When reading a long and complex document (e.g., a textbook), a human reader will often find that they need to refer back to previously introduced material from earlier in the document. Similarly, when reading a news article or tweet about current events, a reader may want to search through other trusted articles to find evidence for particular claims. However, this process of manual searching is inconvenient and disrupts the reading flow, so readers will often refrain from immediately searching for helpful context or evidence. The idea of this project is to build a system that would assist readers in finding such context and evidence seamlessly. A user can query the system simply by highlighting a span of text in a document they are reading. Based on this, our system will automatically generate questions or claims related to the highlighted span that the user might wish to have answered/verified. It will then search for other passages from the document or related documents that would help answer these questions or provide evidence about these claims.
    • Student Responsibilities: Implementing a question generation and passage retrieval pipeline using python, pytorch, and other existing open-source tools for deep learning; Reading academic papers related to question generation, retrieval, question answering, and fact verification
    • Preferred Majors: Computer Science
    • Prerequisites: Knowledge of python, Coursework in machine learning, Interest in natural language processing and deep learning
  • Generating datasets with large language models
    • Description: Large language models like GPT-3 are powerful deep learning systems that have learned to generate reasonable-sounding natural language text, and have the potential to be used in a very wide variety of language-based applications. One potential use case of
    • Student Responsibilities: Using existing pytorch-based implementations of large language models and running large-scale experiments to generate data with them; Implementing training of smaller models using the generated data; Devising new ideas for filtering data Reading NL
    • Preferred Majors: Computer Science
    • Prerequisites: Knowledge of python, Coursework in machine learning, Interest in natural language processing and deep learning

• Faculty/PI: Phebe Vayanos
• Website: https://cais.usc.edu/
• Research Overview:The USC Center for AI in Society is a joint venture between the USC Suzanne Dworak-Peck School of Social Work and the USC Viterbi School of Engineering. The primary goal of USC CAIS is to develop, test, iterate, and demonstrate how AI can be used to tackle the most difficult societal problems. We believe that this can best be achieved by a genuine partnership between computer science, operations research, social work, and community organizations.
• Summer Projects
  • AI for Homelessness
    • Project Description: The goal of this project is to use historical data to build machine learning models that can predict vulnerability in people experiencing homelessness and evaluate existing scoring rules. Click here for more information about this project.
    • Student Responsibilities:  Students are expected to participate actively under a PhD student mentor and faculty PI. Potential for a long-term research position funded by the NSF Research Experience for Undergraduates Program is available.
    • Preferred Majors:  Computer Science, Industrial & Systems Engineering
    • Prerequisites and Preferred Skillsets: strong coding skills in Python required (with prior experience training/evaluating machine learning models in Python), knowledge on optimization (e.g. ISE 330) preferred, coursework in machine learning preferred, prior experience working with real-world and tabular data preferred
  • AI for Biodiversity Conservation
    • Project Description: The goal of this project is to use historical data to build ML models that can predict where jaguars are more likely to be located in Central & South Africa to assist in their conservation. Click here or more information about this project.
    • Student Responsibilities: Students are expected to participate actively under a PhD student mentor and faculty PI. Potential for a long-term research position funded by the NSF Research Experience for Undergraduates Program is available.
    • Preferred Majors: Computer Science, Industrial & Systems Engineering
    • Prerequisites and Preferred Skillsets: strong coding skills in Python required (with prior experience training/evaluating machine learning models in Python), knowledge on optimization (e.g. ISE 330) preferred, coursework in machine learning preferred, prior experience working with real-world and tabular data preferred

• Faculty/PI: Jyotirmoy (Jyo) Deshmukh
• Website: https://cps-vida.github.io/
• Research Overview:Our group’s research interests lie in the intersection of formal methods, control theory, cyber-physical systems, and artificial intelligence. This includes some of the following: 1) Verification of Autonomous CPS, 2) Specification Languages for CPS, 3)Monitoring Security of CPS, 4) Verifiable and Safe Controller Synthesis for Autonomy in CPS, 4) Probabilistic Reasoning about Time-Series Data using Temporal Logic.
• Summer Project
  • Perception and Cognition-aware Decision-Making
  • Project Description: Consider a person who is the backup safety driver for an autonomous vehicle, a drone pilot that simultaneously controls several drones, or a factory worker responsible for monitoring manufacturing robots. Each of these systems is a safety-critical system where the human operator is responsible for the overall safety of the system. In this project, we are interested in exploring if various AI methods can assist a human operator in enhancing the safety of the system. There is considerable recent interest in monitoring the state of a human actor in such a safety-critical human-autonomous system, and this project will primarily focus on developing algorithms for this task. State-of-the-art perception algorithms based on deep neural networks for object recognition, trajectory prediction, etc. could be adapted to perceive the joint state of the human-autonomous system, and part of this project will involve collecting the required data, annotating it (if necessary), and training learning-enabled components to perform monitoring. The project will also involve training decision-making algorithms that use the perceived state of the system to take appropriate safe decisions.
  • Student Responsibilities:  The student will have the following responsibilities:
    • Perform and conduct experiments on a driving simulator
    • Train perception algorithms from multi-modal sensory data
    • Train and research decision-making algorithms based on reinforcement learning and
  • Preferred Majors:  Aerospace & Mechanical Engineering,Astronautical Engineering,Computer Science,Electrical & Computer Engineering,Industrial & Systems Engineering
  • Prerequisites and Preferred Skillsets: Python, Deep Learning, Cognitive science

• Faculty/PI: Jesse Thomason
• Website: http://glamor.rocks/
• Research Overview: The GLAMOR lab brings together natural language processing and robotics to connect language to the world (RoboNLP). Our lab is broadly interested in connecting language to agent perception and action, and lifelong learning through interaction.
• Summer Projects
  • Learning with Language, Perception, and Action
    • Project Description: Language communication reflects interlocutors' reasoning and internal world models. This project will involve jointly learning models with language, world perception, and physical action to enable end-to-end agent behavior and improve continual learning. For example, we may explore ways to train language-guided AI agents and developed multimodal continual learning algorithms that jointly reason over language and the surrounding context in which it was uttered. For inherently multimodal tasks like Vision-and-Language Navigation (VLN) and Visual Question Answering (VQA), late fusion and feature engineering methods have largely given way to large-scale, end-to-end training. We will investigate how to scale such end-to-end approaches to complex settings like dialogue-based turn and action taking, as well as shifts in the underlying task or even modalities themselves through the lifetime of a model's deployment.
    • Student Responsibilities:  Reading research papers, past and present, to understand the state of the art and how we got here. Substantial coding and engineering. Developing research hypotheses and proposing, setting up, and executing experiments to test those hypotheses.
    • Preferred Majors:  Computer Science
    • Prerequisites and Preferred Skillsets: Python, PyTorch, core CS background, interest in language and linguistics
  • Neurosymbolic Reasoning for Language, Vision, and Robotics
    • Project Description: Large language models (LLMs) are commonly used like hammers, with every problem retrofitted as a nail. This project will investigate ways to take advantage of the extra-textual visual world and embodied context in which language is uttered to improve reasoning in language-and-vision and language-guided robotics tasks. GLAMOR has been studying the ways we can incorporate formal structure and task specification into difficult multimodal problems. LLMs are capable of generating blocks of code and formal specification language of quality that was far-fetched just a few years ago. This project will investigate how LLMs can be used to predict semantic structures for question answering tasks, as well as for generating program structures and API calls for virtual agent and physical robot tasks.
    • Student Responsibilities: Reading research papers, past and present, to understand the state of the art and how we got here. Substantial coding and engineering. Developing research hypotheses and proposing, setting up, and executing experiments to test those hypotheses.
    • Preferred Majors: Computer Science
    • Prerequisites and Preferred Skillsets: Python, PyTorch, core CS background, interest in language and linguistics
  • Language Processing for Accessibility and Health
    • Project Description: Natural language is human communication: speech, signs, gestures, glances, nods, and more. This project may work to improve speech and sign recognition by leveraging contextual and structural information, or to apply language technologies to accessibility and health applications. Most language processing technologies are focused on English text as an approximation of natural language. This project will explore a new thread within the general space of language grounding: using contextual and structural information to inform better computational understanding of -natural- spoken and signed language.
    • Student Responsibilities: Reading research papers, past and present, to understand the state of the art and how we got here. Substantial coding and engineering. Developing research hypotheses and proposing, setting up, and executing experiments to test those hypotheses.
    • Preferred Majors: Computer Science
    • Prerequisites and Preferred Skillsets: Python, PyTorch, core CS background, interest in language and linguistics

• Faculty/PI: Heather Culbertson
• Website: https://sites.usc.edu/culbertson/
• Research Overview: The Haptics Robotics and Virtual Interaction (HaRVI) Laboratory explores how humans interact with our world, robots, and technology through touch. The goal of our research is to create natural and intuitive interactions that realistically mimic the touch sensations experienced during interactions with the physical world. We design novel haptic hardware and rendering algorithms to improve the usability of technology, increasing people’s social connectedness, ability to complete specific tasks, and immersiveness in virtual reality. Our research has a strong focus on integrating human perception into all steps of the design process
• Summer Projects:
  • Haptics for Virtual Reality
    • Description: This project focuses on the design, building, and control of haptic devices for virtual reality. Current VR systems lack any touch feedback, providing only visual and auditory information to the user. However, touch is a critical component for our interactions with the physical world and with other people. This research will investigate how we use our sense of touch to communicate with the physical world and use this knowledge to design haptic devices and rendering systems that allow users to interact with and communicate through the virtual world. To accomplish this, the project will integrate electronics, mechanical design, programming, and human perception to build and program a device to display artificial touch sensations to a user with the goal of creating a natural and realistic interaction.
    • Student Responsibilities: Reading and analyzing research papers. Designing, prototyping, and programming haptic devices. Conducting human subject studies.
    • Preferred Majors: Aerospace & Mechanical Engineering, Computer Science
    • Preferred Skillsets: Programming (C++ preferred), CAD/3D printing, electronics and circuits (helpful, but not required)
  • Feedback control for a robotic walker with human user
    • Project Description: Feedback control for a robotic walker with human user
    • Student Responsibilities: Reading and analyzing research papers. Designing, prototyping, and programming a physical system. Conducting human subject studies.
    • Preferred Majors: Aerospace & Mechanical Engineering,Biomedical Engineering,Computer Science
    • Prerequisites and Preferred Skillsets: Programming (C++ preferred), CAD/3D printing, electronics and circuits (helpful, but not required)

• Faculty/PI: Maja Mataric
• Website: https://uscinteractionlab.web.app
• Research Overview: The Interaction Lab focuses on developing computational principles, techniques, models, and interventions to enable socially assistive human-robot interaction that supports human health and wellness. Socially Assistive Robotics (SAR) uses non-contact social interaction interventions involving social, emotional, cognitive, and physical abilities of users, toward improved wellness, communication, learning, and autonomy. The Interaction Lab conducts research in an interdisciplinary SAR arena that uses computing, engineering, and human user studies to characterize, model, and understand complex human behavior. Our projects aim to contribute insights and individualized tools toward mitigating pervasive societal challenges -- including skill training for autism, anxiety, and depression coping, rehabilitation, and healthy aging -- that require sustained personalized support that supplement the efforts of caregivers, clinicians, parents, and educators.
• Summer Projects
  • Harnessing the power of novelty to encourage adherence in long-term human-robot interaction scenarios
    • Project Description: In the field of human-robot interaction, results from user studies are often biased by users' initial interest in the new technology. This phenomenon is referred to as the novelty effect, and it's something researchers typically try to avoid when designing user studies. However, our objective is to use novelty to our advantage and prolong novelty effects by slowly introducing new features to the robot across many repeated interactions. In this project, students will incorporate different modes of interaction (e.g. touch, speech) and positive rewards (e.g. visual light displays, opportunities to customize the robot) on a simple, low-cost robot, in an effort to increase users’ adherence to regular study sessions. We will evaluate these new capabilities by measuring their effect on students' attitude towards the robot and their engagement in study sessions with the robot.
    • Student Responsibilities:  CAD modeling, 3d printing, arduino engineering and programming, Unity game development, optional: creative textile fabrication and crafting
    • Preferred Majors:  Aerospace & Mechanical Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Interest in programming and engineering
  • Understanding and evaluating the ADRD Screening Tool design
    • Project Description: Over six million Americans are living with Alzheimer's disease or another form of dementia, and the rates are projected to increase to nearly 12 million by 2050. The development of accessible non-invasive screening tools is of paramount importance for individuals, family members, and caregivers. We introduce a novel screening tool that includes speech audio and transcripts, eye gaze, and pen pressure across several tasks used for standard dementia screening. We will evaluate our current design based on the capabilities of our ML models to generate accurate predictions.
    • Student Responsibilities: Multimodal Machine Learning Models understanding and programming
    • Preferred Majors: Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: interest in programming and engineering
  • Analyzing emotional labor in human interactions
    • Project Description: This project focuses on using computational methods to understand and predict emotional labor in human-robot and human-human interactions. Students working on this project will be involved in collecting, processing, and analyzing data from real-life interactions. This work will inform the development of socially assistive robots, systems capable of aiding people through social interactions that combine monitoring, coaching, motivation, and companionship. To address the inherently multidisciplinary challenges of this research, the work draws on theories, models, and collaborations from neuroscience, cognitive science, social science, health sciences, and education.
    • Student Responsibilities: data analysis and cleaning, video data extraction
    • Preferred Majors: Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: interest in programming and engineering
  • Helping resolve interpersonal conflict using computational tools
    • Project Description: Working towards utilizing socially assistive robots to help with interpersonal conflict between two or more people, I am looking to investigate key conflict resolution strategies and how computational tools may be able to help with reinforcing these skills. I would eventually like to work with conflict resolution for children.
    • Student Responsibilities: data collection, data analysis and cleaning, literature review
    • Preferred Majors: Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: interest in programming and engineering

• Faculty/PI: Jieyu Zhao
• Website: https://jyzhao.net/lab.html
• Research Overview:Our team's primary focus is on creating trustworthy NLP models. We meticulously investigate the ethical consequences and broader societal effects of NLP models, striving to ensure that language technologies are constructed and employed in ways that align with ethical guidelines and uphold human values.
• Summer Projects
  • Does debias method work for all groups?
    • Project Description: Analyze whether current debias method (e.g. gender bias towards to occupations) only work for certain occupations and sacrifice some low-frequent occupations. If this happens, how can we improve the current debias method.
    • Student Responsibilities:  Read research papers, Writing codes to testify the research ideas, Meet with the mentors and present the work
    • Preferred Majors:  Computer Science
    • Prerequisites and Preferred Skillsets: Have taken ML and NLP classes; Have proficient programming skills.

• • Faculty/PI: Erdem Biyik
  • Website: https://liralab.usc.edu/
  • Research Overview: USC Learning and Interactive Robot Autonomy Lab (LiraLab) develops algorithms for robot learning, safe and efficient human-robot interaction and multi-agent systems. Our mission is to equip robots, or more generally agents powered with artificial intelligence (AI), with the capabilities that will enable them to intelligently learn, adapt to, and influence the humans and other AI agents.
  • Summer Projects
    • Imitation learning from control-constrained demonstrations
      • Project Description: This project will explore efficient ways of performing imitation learning and/or inverse reinforcement learning when the expert demonstrations come from a constrained control interfaces, e.g. due to the controller itself or the suboptimality of the expert human. The applications include tabletop manipulation and autonomous driving.
      • Preferred Majors: Computer Science, Electrical & Computer Engineering
      • Prerequisites and Preferred Skillsets: Python programming language; Fundamentals of machine learning; Some machine learning libraries in Python
    • Self-supervised improvements over reinforcement learning with large pre-trained models
      • Project Description: This project will explore the use of large pre-trained models (e.g., LLMs, VLMs, VQAs, etc.) for creating a self-supervision signal in reinforcement learning. The applications include, but are not limited to, tabletop manipulation.
      • Preferred Majors: Computer Science, Electrical & Computer Engineering
      • Prerequisites and Preferred Skillsets: Python programming language; Basic machine learning knowledge; Familiarity with machine learning libraries in Python
    • Active querying for reinforcement learning from human feedback
      • Project Description: The current implementations of reinforcement learning from human feedback (RLHF) follows the learned policy to generate new queries for the human. In this project, we will explore alternative ways to do it to improve data-efficiency of training. It will involve implementation of active learning techniques.
      • Preferred Majors: Computer Science, Electrical & Computer Engineering
      • Prerequisites and Preferred Skillsets: Python programming language; Basics of machine learning and reinforcement learning; Familiarity with machine learning libraries in Python; Basics of information theory
    • Developing testbeds for reinforcement learning
      • Project Description: This project is about creating testbeds that offer plug-and-play feature for reinforcement learning. These testbeds will be for the other projects in the lab that are about domain transfer in reinforcement learning, reinforcement learning with human feedback, etc. Depending on the needs, the environments may be simple OpenAI Gym environments or more sophisticated algorithms built on Unity, MuJoCo, etc.
      • Preferred Majors: Computer Science, Electrical & Computer Engineering

• Faculty/PI: Yan Liu
• Website: https://melady.usc.edu/
• Research Overview: The USC Melady Lab develops machine learning and data mining algorithms for solving problems involving data with special structure, including time series, spatiotemporal data, and relational data. We work closely with domain experts to solve challenging problems and make significant impacts in computational biology, healthcare, social media analysis, climate modeling, and business intelligence.
• Summer Projects
  • Machine Learning to Guide Ovarian Cancer Treatment
    • Project Description: This project is concerned with decision support technology to inform personalized cancer treatment and improve patient outcomes. The idea is to identify patients who do not respond to prescribed therapies. People with ovarian cancer typically present at an advanced stage, and the standard of care is extensive surgery followed by platinum-based chemotherapy. Unfortunately, approximately 15% of people will be platinum-resistant, meaning that they will relapse within six months of their surgery. This project aims to help such patients by developing a machine learning framework using histopathology images, patient clinical variables, and omics data. The current project consists of three modules of machine learning for building models to process multimodal data, data efficiency via transfer learning and domain adaptation, and interpretability and explainability. We work closely with gynecologic pathologists in Keck School of Medicine.
    • Student Responsibilities:  Collaboratively coding, troubleshooting, and evaluating machine learning models; Recreating state-of-the-art results in the field of machine learning; Literature review; Preprocessing and managing relevant datasets; Visualizing and reporting results.
    • Preferred Majors:  Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: If you have a strong interest in machine learning for health and healthcare applications and proficiency programming with Python, this project is a great place for you. High motivation, attention to detail, and strong organizational skills are preferred.
  • Foundation Models for Time Series Modeling
    • Project Description: Time series data is ubiquitous in domains such as healthcare, finance, traffic, and so on. Effective analysis of time series data is critical for tasks like medical diagnosis, anomaly detection, forecasting and decision-making. However, each time series task typically requires specialized algorithms and labeled data, limiting the ability to handle diverse time series problems jointly. Recently, foundation models have revolutionized natural language processing (NLP) by enabling strong performance on diverse NLP tasks using a single pre-trained model, such as GPT-3, LLaMA and ChatGPT, etc. However, directly applying this approach to time series analysis poses unique challenges. A major challenge is the lack of large datasets required to pre-train deep models on time series as existing collections are much smaller than datasets for NLP and computer vision (CV) tasks. Moreover, time series signals lack the visual cues and natural language descriptions that enable scaling models through multimodal learning. Unlike words or objects, temporal relationships and the evolution of patterns over time are core characteristics of time series. Here, we aim to address those challenges through a novel time series foundation model that can transfer powerful representation learning from LMs by developing methods to project time series into an interpretable embedding space while accounting for their temporal properties. The key goals of a time series foundation model can be summarized as: (1) Provide a unified framework for handling diverse time series tasks like classification, anomaly detection and forecasting using a single model. (2) Allow analyzing time series data effectively even when labeled examples are limited, by leveraging the transfer learning capabilities from language model pre-training. (3) Learn interpretable embeddings that reveal intrinsic patterns and mechanisms underlying time series data. Expected outcomes: The research aims to deliver several impactful outcomes. A key goal is to benchmark our time series foundation model on public evaluation datasets and leaderboards, comparing its general framework performance to specialized state-of-the-art models on tasks like forecasting and anomaly detection. The pre-trained model and an open-source library implementing foundational time series analytics routines using it will also be publicly released to facilitate broader adoption. Additionally, we expect to expand the collection of multi-modal time series datasets augmented with textual side information, which will be valuable resources for future work exploring the integration of temporal and language data.
    • Student Responsibilities: Communicating, collaboratively coding, troubleshooting, and evaluating machine learning models; Recreating state-of-the-art results in the field of machine learning; Literature review; Preprocessing and managing relevant datasets; Visualizing and reporting results.
    • Preferred Majors: Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: If you have a strong interest in large pre-trained NLP modeling or machine learning for time series applications and proficiency in programming with Python, this project is a great place for you. High motivation, attention to detail, and strong organizational skills are preferred. This is a computer science project, and does not require any background in biology and medicine but some is a plus.

• Faculty/PI: Vatsal Sharan
• Website: https://vatsalsharan.github.io/
• Research Overview: I work on machine learning and theoretical computer science. My research aims to understand how to solve learning and inference tasks in the face of various computational and statistical constraints, such as limited memory or too little data. I am also broadly interested in the desiderata in practical applications beyond these constraints as well---particularly in notions of robustness. My interests are both in developing theoretical frameworks to understand the fundamental limits in these modern settings, and in designing practical algorithms to tackle these problems.
• Summer Projects:
  • Machine Learning Mysteries
  • Description: Students will explore foundational questions regarding machine learning, using a combination of systematic experiments and theoretical analysis. Both students interested in performing systematic experiments to tease out phenomenon in practice, and those interested in using theoretical tools to prove novel guarantees are welcome. The exact questions and their scope is broad, but the following are some potential options: 1. Understanding deep learning: One particular question of interest here is to understand why neural networks generalize despite having the capacity to overfit. We've been exploring what role the data itself has to play in this mystery, exploring connections to the amazing self-supervised learning capability of neural networks in the process. 2. Data augmentation and amplification: In some recent work ("Sample Amplification"), we showed that it is often possible to generate new samples from a distribution without even learning it. We will explore how this ties into various data augmentation techniques, and develop new frameworks to increase dataset size. 3. Fairness and robustness: We will explore how to train models which are robust in many ways, such as to changes in the data distribution, or to do well on minority sub-populations (and not just on average over the entire data). 4. Computational-statistical tradeoffs: This is a more theoretical direction, to understand when computational efficiency might be at odds with statistical requirements (the data needed to learn). Recent work has opened by much uncharted territory, particularly with respect to the role of memory in learning, which we will explore.
  • Preferred Majors: Computer Science, Electrical & Computer Engineering
  • Preferred Skillsets: Familiarity with probability, linear algebra, calculus, and analysis of algorithms. Some basic understanding of machine learning would be very helpful for certain projects.

• Faculty/PIs: Harsha Madhyastha, Ramesh Govindan, Seo Jin Park
• Website: https://nsl.usc.edu/
• Research Overview: At NSL, we conduct research on the design and implementation of a wide range of networked computing systems. Examples of current topics of focus include augmented/mixed reality, web archival, computing for sustainability, and serverless computing.
• Summer Projects
  • Reviving Dead Links on the Web (Led by Dr. Harsha Madhyastha)
    • Project Description: A key feature of the web is that the links on any page enable visitors to discover other related pages. One would expect any link to break only when the page it is pointing to no longer exists. Instead, the web is littered with millions of broken links today merely because, when a website is reorganized and the URLs for its pages change, links which specify the old URLs for these pages often cease to work. To ensure that the effort that has gone into creating appropriate links all across the web does not go to waste, we have developed a system called FABLE. Given any broken link, FABLE helps resurrect that link by discovering the linked page's new URL, if the page still exists on the web. The goal of this summer project will be to make the utility of FABLE broadly available by making FABLE a cloud-hosted service that can be used to fix broken links on the pages of any website.
    • Student Responsibilities:
      • Deploy FABLE in a cloud service such as AWS or Microsoft Azure
      • Crawl pages from many sites on the web, find the broken links they include, and run FABLE to find the corresponding new URLs
      • Reach out to the administrators of crawled websites to figur
    • Preferred Majors:  Computer Science
    • Prerequisites and Preferred Skillsets: Strong experience with web development, both front-end and back-end; Experience using cloud services such as Amazon Web Service, Microsoft Azure, or Google Cloud; Experience dealing with large datasets
  • 3D Video Analytics (led by Dr. Ramesh Govindan)
    • Project Description: The rise of Augmented/Virtual Reality (AR/VR) has paved the way for next-generation immersive 3D media. This creates new research opportunities to scale 2D computer vision technology to 3D. We want to investigate how to apply computer vision models to perform tasks like object detection, tracking, etc., on 3D video from a computer systems perspective. We are looking for undergraduate student(s) who will work in NSL (Networked Systems Laboratory) to build research prototypes and push the state-of-the-art.
    • Student Responsibilities:  Students will be exposed to cutting-edge tools for research and development in the intersection of AR/VR and Computer Vision. They will be responsible for the following:
      • Apply 3D computer vision models on 3D video.
      • Develop algorithms to optimize for latency and throughput of 3D analytics.
      • Work with Ph.D. students on an existing codebase to integrate the above algorithms.
      • Read existing literature in related areas.
    • Preferred Majors:  Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: These are the preferred (not necessary) skillsets that we are looking for:
      • Programming in C++ and Python.
      • Machine learning frameworks like Pytorch or Tensorflow.
      • Some experience with 3D data formats like point clouds and/or meshes.
      • XR development
  • Flash burst computing (led by Dr. Seo Jin Park)
    • Project Description: Flash burst computing is motivated by two growing trends: big data and cloud computing. Today, many businesses and web services store staggering quantities of data in the cloud and lease relatively small clusters of instances to run analytics queries, train machine learning models, and more. However, the exponential data growth, combined with the slowdown of Moore's law, makes it challenging (if not impossible) to run such big data processing tasks in real-time. Most applications run big data workloads on timescales of several minutes or hours, and resort to complex, application-specific optimizations to reduce the amount of data processing required for interactive queries. This design pattern hinders developer productivity and restricts the scope of applications that can use big data.

      My research aims to enable interactive, cost-effective big data processing through "flash bursts." Flash bursts enable an application to use a large portion of a shared cluster for short periods of time. This could allow big data applications to complete significantly faster, with cost comparable to leasing a few instances for a longer period of time. A flash-burst-capable cloud could enable new real-time applications that use big data in ad hoc ways, without relying on query prediction and precomputation; e.g., a security intrusion detection system could analyze large-scale logs from many machines in real-time as threats emerge. It could also improve developer productivity (e.g., a machine learning researcher could train models and iterate on new ideas quickly). Finally, flash bursts could reduce costs, since applications would no longer need to precompute results for all potential queries.

    • Student Responsibilities: Measure the performance of various distributed computing applications, such as data analytics, with varying scale. Find the scalability bottleneck for those applications and fix them to enable interactive data-intensive computing.
    • Preferred Majors:  Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Strong C++ hacking skills, experience with data analytics frameworks (e.g., Spark) or distributed training on GPU.

• Faculty/PI: Chao Wang
• Website: https://sites.usc.edu/chaowang/prospective-students/
• Research Overview:We develop mathematically rigorous methods and tools to ensure that critical software components (including those based on machine learning) satisfy both functional and nonfunctional properties (e.g., safety, security and fairness properties).
• Summer Projects
  • Using machine learning to improve software verification
    • Project Description: The objective of this project is to identify and improve some of the most challenging tasks in software verification by levering machine learning techniques.
    • Student Responsibilities:  Students are expected to help PhD students collecting benchmark programs and running experimental evaluation.
    • Preferred Majors:  Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Suitable for junior and senior students in computer science or computer engineering.
  • Verifying fairness and data bias robustness of machine learning models
    • Project Description: Verifying fairness and data bias robustness of machine learning models
    • Student Responsibilities: Students are expected to help PhD students by collecting benchmark datasets and running experiments.
    • Preferred Majors: Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Suitable for junior and senior students in computer science or computer engineering

• Faculty/PI: Daniel Seita
• Website: https://slurm-lab-usc.github.io/
• Research Overview: The Sensing, Learning, and Understanding for Robotic Manipulation (SLURM) Lab at the University of Southern California studies how we can enable robots to interact with and manipulate geometrically and perceptually challenging objects. Our long-term goal is to develop robots that can reliably manipulate objects in less-structured and messier real world settings. To this end, we develop novel methods for improving robotic manipulation by incorporating state of the art machine learning algorithms, perception methods, and hardware platforms. Professor Daniel Seita directs the SLURM Lab.
• Summer Project
  • Manipulating Mixtures of Liquids and Solids
    • Project Description: Humans use spoons and similar tools to transport and manipulate liquids and solids. For example, we might use spoons to transfer sauces from one bowl to another when cooking, or use spoons to retrieve food items when eating soup, stew, breakfast cereal, chopped almonds, etc. We wish to train robots to similarly manipulate tools in these contexts, with generalization to different mixtures of liquids and solids.To start, we may build upon a simulation environment that can simulate mixtures of liquids and solids. Long-term, we hope that this work can lead the groundwork for robotic assistive feeding, and we aim to collaborate with other labs to work on physical feeding experiments.
    • Student Responsibilities:  The student would be developing and using a simulator for simulating complex deformables and their interactions. A stretch goal would be to show applicability on a physical robot system (or to collaborate with a team which does this).
    • Preferred Majors:  Computer Science
    • Prerequisites and Preferred Skillsets: The student would need to be familiar with Python and the basics of machine learning (or be willing to learn as they work on the project). Past experience in robotics is not required.

• Faculty/PI: Yue Wang
• Website: https://yuewang.xyz
• Research Overview:My research lies in the intersection of computer vision, computer graphics, and robotics. My goal is to use machine learning to enable robotic intelligence with minimal human supervision. I study how to design 3D learning systems which leverage geometry, appearance, motion, and any other cues that are naturally available in sensory inputs. I am also broadly interested in fundamental deep learning tools and eclectic applications on top of these systems.
• Summer Project
  • Generative Neural Scene Representations
    • Project Description: Neural fields, also known as coordinate-based neural networks, have emerged as innovative representations for a variety of signals, including 3D geometry, images, and sound. These representations have demonstrated significant progress in addressing computer vision and graphics challenges such as novel view synthesis and scene reconstruction. Despite these advancements, the implementation of neural fields in robotics remains in its nascent stages. The central question of our study is whether these innovative representations can inform continuous decision-making in autonomous driving. Although recent efforts have provided promising evidence, numerous modeling and computational obstacles continue to impede the development of stable and efficient neural field methods. In this project, our objective is to address many challenges by investigating generative neural voxels, a distinct hybrid representation stemming from neural field training. Neural voxels seamlessly integrate geometric and semantic information into latent features, and, unlike binary occupancy, they also encompass semantic features derived from a 2D foundation model. Generated through NeRF training, multi-view images are encoded by an image encoder and subsequently processed by a 2D-to-3D lifting module, transforming 2D feature maps into 3D feature voxels or neural voxels. These 3D feature voxels are then decoded via volumetric rendering, producing RGB images, depth maps, and 2D semantic features. The resulting prediction targets are supervised using ground-truth images, LiDAR sparse depths, and features extracted from 2D foundation models.
    • Student Responsibilities:  Fundamental Research
    • Preferred Majors:  Aerospace & Mechanical Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Computer vision, machine learning, robotics

• Faculty/PI:Christopher Torng
• Website: https://sites.usc.edu/acorn-research/ 
• Summer Projects
  • Enabling Rapid Chip Design with Agile Flow Tools
    • Project Description: Achieving high code reuse in physical design flows is challenging but increasingly necessary to build complex systems. We present a vision and framework based on modular flow generators that encapsulates coarse-grained and fine-grained reusable code in modular nodes and assembles them into complete flows. These agile flow tools are being designed to enable students across the country in R1/R2 universities to successfully build chips in advanced technologies.
    • Student Responsibilities: Depending on the student's interests and abilities, the candidate can work with physical design flow tools using Tcl and Python, or experiment with designing small hardware blocks pushed from RTL to GDS in a commercial digital ASIC tool flow.
    • Preferred Majors: Electrical and Computer Engineering
    • Prerequisites and Preferred Skillsets: Candidates should: - Have initiative and curiosity - Work well in teams - Have prior programming experience in Python - Have prior programming experience in any hardware description language (e.g., Verilog) - Have taken some VLSI coursework
  • Rapid Runtime Reconfigurable Arrays for Wideband Spectrum Sensing and Machine Learning
    
    • Project Description: Commercial and military demands on the electromagnetic spectrum are driving RF systems to operate in increasingly congested and complex environments. These systems must analyze large volumes of continuously streaming data, detect and characterize waveforms, and wake up downstream decision-making applications, all within unknown environments. We will build and ask fundamental questions about how to build compute-dense runtime reconfigurable arrays with fast and flexible program switching controlled by embedded real-time schedulers.
    • Student Responsibilities: Depending on the student's interests and abilities, the candidate can design hardware blocks in Verilog, write flow scripts in Python/Tcl, and more generally have an opportunity to explore compiler-level work (how to run software on this accelerator) or VLSI-level work (how to build a chip for this accelerator)
    • Preferred Majors: Electrical and Computer Engineering
    • Prerequisites and Preferred Skillsets: Candidates should: - Have initiative and curiosity - Work well in teams - Have prior programming experience in Python - Have prior programming experience in any hardware description language (e.g., Verilog)
  • Specializing Communication at the Inter-Chiplet Boundary for Energy-Efficient Machine Learning
    • Project Description: An open chiplet ecosystem would allow heterogeneous integration of chips in a compact, advanced package. Building a chiplet ecosystem represents a tremendous paradigm shift in the time and cost of assembling future computing systems and constitutes a key thrust in the CHIPS and Science Act research strategy. How can we build chiplet-based systems to be simpler, faster, and more efficient when domain-specific hardware accelerators are communicating across inter-chiplet interfaces?
    • Student Responsibilities: Depending on the student's interests and abilities, the candidate can design hardware blocks in Verilog, write flow scripts in Python/Tcl, and more generally have an opportunity to explore compiler-level work (how to run software on this accelerator) or VLSI-level work (how to build a chip for this accelerator)
    • Preferred Majors: Electrical and Computer Engineering
    • Prerequisites and Preferred Skillsets: Candidates should: - Have initiative and curiosity - Work well in teams - Have prior programming experience in Python - Have prior programming experience in any hardware description language (e.g., Verilog)

• Faculty/PI: Justin Haldar
• Website: https://mr.usc.edu/
• Summer Project
  • Project Description: Magnetic resonance imaging (MRI) technologies provide unique capabilities to probe the mysteries of biological systems, and have enabled novel insights into anatomy, metabolism, and physiology in both health and disease. However, while MRI is decades old, is associated with multiple Nobel prizes (in physics, chemistry, and medicine), and has already revolutionized fields like medicine and neuroscience, current MRI methods are still very far from achieving the full potential of the MRI signal. Specifically, modern MRI methods suffer due to long data acquisition times, limited signal-to-noise ratio, high monetary costs, and various other practical and experimental limitations — this limits the amount of information we can extract from living human subjects, and often precludes the use of advanced experimental methods that could otherwise increase our understanding by orders-of-magnitude. Our research group addresses such limitations from a signal processing perspective, developing novel methods for data acquisition, image reconstruction, and parameter estimation that combine: (1) the modeling and manipulation of physical imaging processes; (2) the use of novel constrained signal and image models; (3) novel theory to characterize signal estimation frameworks; and (4) fast computational algorithms and hardware. Methods we developed have enabled substantial acceleration of routine modern MRI exams, and have also enabled the development of highly-informative next-generation MRI experiments that were previously impractical. Our approaches are often based on jointly designing data acquisition and image reconstruction methods to exploit the inherent structure that can be found within high-dimensional data, and we do our best to take full advantage of the "blessings of dimensionality" while mitigating the associated "curses."
  • Preferred Majors:  Electrical & Computer Engineering
  • Prerequisites and Preferred Skillsets: Signal processing, linear algebra, programming

• Faculty/PI: Stephen Cronin
• Website: https://cronin-lab.usc.edu/
• Research Overview:We focus  on a broad range of interrelated topics in physics, chemistry, materials science and nanotechnology
• Summer Projects
  • Enhanced Combustion of Carbon-free ("Green") Fuels using Nanosecond High Voltage Plasma Discharge
    • Project Description: Hydrogen (H2) and ammonia (NH3) are considered as the leading candidates among all green fuels. While the combustion of the carbon-free fuels doesn't produce ay CO2, the are difficult to burn and produce a significant amount of nitric oxide (NO and NO2). This project explores the enhanced combustion of these carbon-free fuels using a transient plasma ignition system, which is based on USC patented technology of high voltage nanosecond pulse discharges.
    • Student Responsibilities:  Students involved in this project will run various engine tests with and without the transient plasma ignition system and measure the exhaust gas characteristics (i.e., CO, CO2, NO, and NO2).
    • Preferred Majors:  Aerospace & Mechanical Engineering, Astronautical Engineering, Chemical Engineering, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: No prior skills needed.
  • Solar Fuel Production using Semiconductor Photocatalysts
    • Project Description: Solar Fuel Production using Semiconductor Photocatalysts
    • Student Responsibilities: Students involved in this project will build and test various semiconductor photocatalysts including silicon, GaAs, and InP semiconductors. Testing will involve characterizing the photo-conversion efficiency using a solar simulator and an electrochemical
    • Preferred Majors: Aerospace & Mechanical Engineering,Astronautical Engineering,Chemical Engineering,Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: no prior skills are needed.

Cyber-Physical Systems

• Faculty/PI: Paul Bogdan
• Website: https://cps.usc.edu/studs.html
• Research Overview: Our research is dedicated to understanding and designing Cyber Physical Systems. CPS is a highly interdisciplinary area and our group works on a variety of topics. In general, we are interested in the fundamental and applied questions relating to the structure and dynamics of networks.

• Faculty/PI: Peter Beerel
• Website: https://sites.usc.edu/eessc/
• Research Overview: The E2S2C group, led by Professor Peter A. Beerel, has active research efforts spanning circuits, micro-architecture, and algorithms that target a variety of emerging areas in energy-efficient, secure, and sustainable computing. The group is guided by academic curiosity, integrity, and the spirit of collaboration to solve real-world problems using the wide array of mathematics that make up the foundation of Electrical and Computer Engineering. The group’s current research projects include topics in machine-learning algorithm hardware co-design, superconducting electronics, hardware security, and asynchronous VLSI design. The group is also collaborating on multidisciplinary problems including mitigating wild-fires using a network of drones.

• Faculty/PI: Viktor Prasanna
• Website: fpga.usc.edu | dslab.usc.edu
• Research Overview:
  • The FPGA/Parallel Computing Lab is focused on solving data, compute and memory intensive problems in the intersection of high speed network processing, data-intensive computing, and high performance computing. We are exploring novel algorithmic optimizations and algorithm-architecture mappings to optimize performance of parallel and heterogeneous architectures including Field-Programmable Gate Arrays (FPGA), general purpose multi-core (CPU) and graphics (GPU) processors.
  • The Data Science Lab focuses on applying machine learning, data mining, and network analysis to real-world problems in society and industry.
• Summer Projects
  • FPGA acceleration of GNN
    • Project Description: Designing and implementing specific GNN models and performance evaluation
    • Student Responsibilities:  Coding using high level languages
    • Preferred Majors: Computer Science, Electrical & Computer Engineering
  • GNNs for SAR ATR
    • Project Description: GNNs for SAR ATR
    • Student Responsibilities: Use ML frameworks for performance evaluation
    • Preferred Majors: Computer Science Electrical & Computer Engineering

• Faculty/PI: Hossein Hashemi
• Website: https://hhlab.usc.edu/
• Research Overview: We analyze, design, and implement integrated circuits and systems for communications, sensing, and imaging applications.
• Specifically, we envision high-performance sensing and communication devices that can be embedded in the environment to increase our awareness, improve the quality of life, and create an intelligent and responsive ambient.
• Summer Projects:
  • Millimeter-Wave Integrated Circuits
    • Description: Modern and future wireless communication systems (5G, 6G), automotive radars, and high-resolution 3D imagers operate at millimeter wave frequencies. This project involves simulations and design of millimeter-wave integrated circuits.
    • Student Responsibilities: The student will design and simulate millimeter-wave integrated circuits.
    • Preferred Majors: Electrical & Computer Engineering
    • Preferred Skillsets: Required background- Basic electromagnetics (Maxwell equations), Basic transistor-level analog circuits; Additional Preferred Knowledge (not necessary) - Cadence design tool, Python programming
  • Silicon Photonics Integrated Circuits
    • Description: The project involves application of silicon photonics integrated circuits for lidar, optical computing, and other applications.
    • Student Responsibilities: The student will work on design and simulations of photonic components and photonic integrated circuits.
    • Preferred Majors: Electrical & Computer Engineering
    • Preferred Skillsets: Required Knowledge - Basic electromagnetic (solving Maxwell equation with boundary value)

• Faculty/PI: Yasser Khan
• Website: khan.usc.edu
• Research Overview: We use additive manufacturing and hardware AI to produce skin-like wearables, implantables, and ingestibles. These medical devices are being used for precision health and psychiatry.
• Summer Projects
  • Wearables for Mental Health
    • Project Description: Developing multi-modal wearables for brain and behavior study involves creating devices that collect diverse data types like physiological and neurological signals. These wearables, used in daily life, help analyze complex aspects of human cognition and behavior, leveraging advanced data analysis for applications in healthcare, research, and technology.
    • Student Responsibilities:
      • Wearable Sensor Design: Designing the sensors that will be integrated into the wearables, ensuring they are suitable for capturing the required data (like physiological and neurological signals).
      • Fabrication: Physically creating the components of wearable devices, which may involve working with different materials and manufacturing techniques.
      • Electronics and Embedded System Programming: Developing the electronic systems that power the wearables, including programming the microcontrollers or embedded systems that manage sensor data collection and device operation.
      • Printed Circuit Board (PCB) Design: Designing PCBs that connect the electronic components of the wearables, including sensors, processors, and power supplies.
      • Experimental Setup: Setting up experiments to test and validate the wearables, which includes preparing the environment, the devices, and any supporting technology.
      • Data Collection: Operating the wearables during experiments to collect data.
        Data Processing: Analyzing the collected data to extract meaningful information, which may involve cleaning the data, performing statistical analyses, and using AI algorithms.
    • Fabrication: Physically creating the components of wearable
    • Preferred Majors:  Aerospace & Mechanical Engineering, Biomedical Engineering, Chemical Engineering, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Electronics design, fabrication experience, coding, embedded systems, prototyping, AI/ML tools use.

• • MRI coils and wearables
    • Project Description: The project involves developing wearable devices and flexible coils specifically designed for use in MRI environments. These advanced tools are capable of concurrently collecting physiological data and MRI images, enabling simultaneous monitoring of physiological responses and detailed imaging of internal body structures. This dual functionality enhances the depth and quality of data collected during MRI scans, providing valuable insights for medical research and diagnosis.
    • Student Responsibilities:
      • Wearable Sensor Design: Designing the sensors that will be integrated into the wearables, ensuring they are suitable for capturing the required data (like physiological and neurological signals).
      • Fabrication: Physically creating the components of the wearable devices, which may involve working with different materials and manufacturing techniques.
      • Electronics and Embedded System Programming: Developing the electronic systems that power the wearables, including programming the microcontrollers or embedded systems that manage sensor data collection and device operation.
      • Printed Circuit Board (PCB) Design: Designing PCBs that connect the electronic components of the wearables, including sensors, processors, and power supplies.
      • Experimental Setup: Setting up experiments to test and validate the wearables, which includes preparing the environment, the devices, and any supporting technology.
      • Data Collection: Operating the wearables during experiments to collect data.
      • Data Processing: Analyzing the collected data to extract meaningful information, which may involve cleaning the data, performing statistical analyses, and using AI algorithms.
    • Preferred Majors: Aerospace & Mechanical Engineering, Biomedical Engineering, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Electronics design, fabrication experience, coding, embedded systems, prototyping, AI/ML tools use.
  • Ingestible electronics
    • Project Description: A pill-sized electronic module is being developed to detect chemicals and gases within the gut. This innovative device is designed for ingestion and aims to analyze the internal chemical environment of the gastrointestinal tract, providing insights into digestive health and potentially aiding in diagnosing various gut-related conditions.
    • Student Responsibilities:
      • Sensor Design: Designing the sensors that will be integrated into the ingestible.
      • Fabrication: Physically creating the components of ingestible devices, which may involve working with different materials and manufacturing techniques.
      • Electronics and Embedded System Programming: Developing the electronic systems that power the ingestible, including programming the microcontrollers or embedded systems that manage sensor data collection and device operation.
      • Printed Circuit Board (PCB) Design: Designing PCBs that connect the electronic components of the ingestible, including sensors, processors, and power supplies.
      • Experimental Setup: Setting up experiments to test and validate the ingestible, which includes preparing the environment, the devices, and any supporting technology.
      • Data Collection: Operating the ingestible during experiments to collect data.
      • Data Processing: Analyzing the collected data to extract meaningful information, which may involve cleaning the data, performing statistical analyses, and using AI algorithms.
    • Preferred Majors: Aerospace & Mechanical Engineering, Biomedical Engineering, Chemical Engineering, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Electronics design, fabrication experience, coding, embedded systems, prototyping, AI/ML tools use.
  • Organic electrochemical-transistor based sensors and circuits
    • Project Description: The objective of this project is to utilize sensors and circuits based on printed organic electrochemical transistors to develop wearables equipped with Artificial Intelligence (AI). These advanced wearables are intended to harness AI capabilities for enhanced performance and functionality in various applications.
    • Student Responsibilities:
      • Wearable Sensor Design: Designing the sensors that will be integrated into the wearables, ensuring they are suitable for capturing the required data (like physiological and neurological signals).
      • Fabrication: Physically creating the components of the wearable devices, which may involve working with different materials and manufacturing techniques.
      • Electronics and Embedded System Programming: Developing the electronic systems that power the wearables, including programming the microcontrollers or embedded systems that manage sensor data collection and device operation.
      • Printed Circuit Board (PCB) Design: Designing PCBs that connect the electronic components of the wearables, including sensors, processors, and power supplies.
      • Experimental Setup: Setting up experiments to test and validate the wearables, which includes preparing the environment, the devices, and any supporting technology.
      • Data Collection: Operating the wearables during experiments to collect data.
      • Data Processing: Analyzing the collected data to extract meaningful information, which may involve cleaning the data, performing statistical analyses, and using AI algorithms.
    • Preferred Majors: Aerospace & Mechanical Engineering, Biomedical Engineering, Chemical Engineering, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets: Electronics design, fabrication experience, coding, embedded systems, prototyping, AI/ML tools use.

• Faculty/PI: Corey Baker
• Website: https://github.com/netreconlab
• Research Overview:The Network Reconnaissance Lab (NetRecon) investigates full stack systems for distributing, protecting, and authenticating data in opportunistic networking scenarios for remote patient monitoring, smart cities, and natural disasters to improve the livelihood of people. We evaluate real-world applications of opportunistic delay tolerant networks (DTNs and human centered computing to empower device-to-device (D2D) social networks for crowd sourcing information. Leveraging opportunistic communication provides complementary solutions to traditional networks which are typically dependent upon centralized infrastructures such as the Internet. The goal of the NetRecon Lab is to make critical data accessible to vulnerable communities in the midst of intermittent and poor connectivity while minimizing delay.
• Summer Project
  • Designing Usable Symptom Monitoring Systems & Apps for Post Cancer Surgery Patients
    
    • Project Description: Current distress screening measures are not designed in a patient-centric manner. For example, patients may misinterpret how to complete a distress questionnaire or symptom report. Involving patients in the design process of a distress screening intervention could allow for an increased perception of benefits by patients, since the design took patient values into consideration. The project is to improve and/or add new usable designs to our iOS app, Assuage, a medical research platform that hosts multiple user interfaces (UIs) for capturing quality of life survey information for patients and displaying outcomes to physicians. Students will design apps on the patient and physician side and assist in IRB approved studies with real patients and doctors.
    • Student Responsibilities:
      • Add code to our GitHub repositories related to iOS and/or watchOS
      • Add new quality of life based surveys to Assuage
      • Design qualitative surveys and interact with patients and doctors
      • Work collaboratively with engineers and researchers from the medical field
      • Analyze data collected from databases and/or surveys
    • Preferred Majors: Computer Science, Computer Engineering, Cognitive Science, or Human Centered Computing related field
    • Prerequisites and Preferred Skillsets: Have an Apple/macOS computer, strong programming experience in: Swift (ideal), C++, Java Script, Python, familiar with Git and Github

• Faculty/PI: Feifei Qian
• Website: https://sites.usc.edu/qian/
• Research Overview: Animals -- lizards, snakes, insects -- often exhibit novel strategies in effectively interacting with their physical environments and generating desired responses for locomotion. In our lab, we are interested in creating robots that can do the same. Our approach integrates engineering, physics, and biology to discover the general principles governing the interactions between bio-inspired robots and their locomotion environments. For example, how do legged animals and robots use solid-like and fluid-like responses from soft sand and mud to produce effective movement? How can insect-like and snake-like robots take advantage of obstacle collisions to navigate within cluttered environments? We use these principles to create novel sensing and control strategies that can allow robots to perceive and intelligently elicit environment responses to achieve desired motion, even from traditional-considered "undesired" environments such as flowing sand, yielding mud, and cluttered obstacle fields.
• Summer Projects:
  • Robot Locomotion and Navigation on Complex Terrains
    • Description: Physical environments can provide a variety of interaction opportunities for robots to exploit towards their locomotion goals. However, it is unclear how to even extract information about - much less exploit - these opportunities from physical properties (e.g., shape, size, distribution) of the environment. This project integrates mechanical engineering, electrical engineering, computer science, and physics, to discover the general principles governing the interactions between bio-inspired robots and their locomotion environments, and uses these principles to create novel control, sensing, and navigation strategies for robots to effectively move through non-flat, non-rigid, complex terrains. For example, with a simple interaction model of robot-obstacle interactions, a bio-inspired multi-legged robot can intelligently exploit obstacle disturbances to generate desired locomotion dynamics. With a better understanding of sand responses to robot leg interactions, we are developing robots with direct-drive or quasi direct-drive legs that can sensitively "feel" the stability and erodibility of desert soil, to help geoscientists collect invaluable measurements on desertification through every step.
    • Student Responsibilities: Selected candidate will assist in development of robotic platforms (Arduino, Raspberry Pi, SolidWorks, 3D Printing), locomotion experiment data collection and analysis (motion capture tracking, force measurements), modelling and simulation (MATLAB, Python, or C++).
    • Preferred Majors: Aerospace & Mechanical Engineering, Astronautical Engineering, Biomedical Engineering, Computer Science, Electrical & Computer Engineering
    • Preferred Skillsets: Mechanical design (SolidWorks) and fabrication (hand and power tools), programming (Python or C++), a good understanding of physics or mechanics. Experiences with Arduino, Rasp Pi, and motors, would be a plus.

• Faculty/PI: Somil Bansal
• Website: https://smlbansal.github.io/sia-lab/
• Research Overview: We seek to design autonomous and robotic systems that operate with guaranteed safety and performance in new settings and environments. In short-term, our research emphasis is on bridging machine learning with classical, model-based planning and control methods to achieve this endeavor.
• Summer Projects:
  • Detecting and Mitigating Anomalies in Vision-Based Controllers
    • Description: Autonomous systems, such as self-driving cars and drones, have made significant strides in recent years by leveraging visual inputs and machine learning for decision-making and control. Despite their impressive performance, these vision-based controllers can make erroneous predictions when faced with novel or out-of-distribution inputs. Such errors can cascade to catastrophic system failures and compromise system safety, as exemplified by recent self-driving car accidents. In this project, we aim to design a run-time anomaly monitor to detect and mitigate such closed-loop, system-level failures. Specifically, we leverage a reachability-based framework to stress-test the vision-based controller offline and mine its system-level failures. This data is then used to train a classifier that is leveraged online to flag inputs that might cause system breakdowns. The anomaly detector highlights issues that transcend individual modules and pertain to the safety of the overall system. We will also design a fallback controller that robustly handles these detected anomalies to preserve system safety. In our preliminary work, we validate the proposed approach on an autonomous aircraft taxiing system that uses a vision-based controller for taxiing. Our results show the efficacy of the proposed approach in identifying and handling system-level anomalies, outperforming methods such as prediction error based detection, and ensembling, thereby enhancing the overall safety and robustness of autonomous systems. Preliminary experiment videos can be found at: https://phoenixrider12.github.io/failure_mitigation 
    • Preferred Majors: Aerospace & Mechanical Engineering, Computer Science, Electrical & Computer Engineering
    • Preferred Skillsets: Good programming skills. Coursework in robotics and/or control. Familiarity with machine learning and deep learning tools (such as PyTorch and Tensorflow) is a plus.
  • Vision-based navigation in new environments
    • Description: Autonomous robot navigation is a fundamental and well-studied problem in robotics. However, developing a fully autonomous robot that can navigate in a priori unknown environments is difficult due to challenges that span dynamics modeling, onboard perception, localization and mapping, trajectory generation, and optimal control. Classical approaches such as the generation of a real-time globally consistent geometric map of the environment are computationally expensive and confounded by texture-less, transparent or shiny objects, or strong ambient lighting. End-to-end learning can avoid map building, but is sample inefficient. Furthermore, end-to-end models tend to be system-specific. In this project, we will explore modular architectures to operate autonomous systems in completely novel environments using onboard perception sensors. These architectures use machine learning for high-level planning based on perceptual information; this high-level plan is then used for low-level planning and control via leveraging classical control-theoretic approaches. This modular approach enables the conjoining of the best of both worlds: autonomous systems learn navigation cues without extensive geometric information, making the model relatively lightweight; the inclusion of the physical system structure in learning reduces sample complexity relative to pure learning approaches. Our preliminary results indicate a 10x improvement in sample complexity for wheeled ground robots. Our hypothesis is that this gap will only increase further as the system dynamics become more complex, such as for an aerial or a legged robot, opening up new avenues for learning navigation policies in robotics. Preliminary experiment videos can be found at: https://smlbansal.github.io/LB-WayPtNav/ and https://smlbansal.github.io/LB-WayPtNav-DH/
    • Preferred Majors: Aerospace & Mechanical Engineering, Astronautical Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites: Some familiarity (through past research or coursework) with machine learning, control theory, or robotics would be desirable (but not necessary); Comfort with coding; Experience with hardware is a plus.
  • Safe assurances for learning and vision-driven robotic systems
    • Description: Machine learning-driven vision and perception components make a core part of the navigation and autonomy stacks for modern robotic systems. On the one hand, they enable robots to make intelligent decisions in cluttered and a priori unknown environments based on what they see. On the other hand, the lack of reliable tools to analyze the failures of learning-based vision models make it challenging to integrate them into safety-critical robotic systems, such as autonomous cars and aerial vehicles. In this project, we will explore designing a robust control-based safety monitor for visual navigation and mobility in unknown environments. Our hypothesis is that rather than directly reasoning about the accuracy of the individual vision components and their effect on the robot safety, we can design a safety monitor for the overall system. This monitor detects safety-critical failures in the overall navigation stack (e.g., due to a vision component itself or its interaction with the downstream components) and provides safe corrective action if necessary. The latter is more tractable because the safety analysis of the overall system can be performed in the state-space of the system, which is generally much lower-dimensional than the high-dimensional raw sensory observations. Preliminary results on simulated and real robots demonstrate that our framework can ensure robot safety in various environments despite the vision component errors (the videos of some of our preliminary experiments can be found at https://smlbansal.github.io/website-safe-navigation/). In this project, we will extend the proposed framework to more complex and high-dimensional robotic systems, such as drones and legged robots. Other than ensuring robot safety, we will also explore using the proposed framework to mine critical failures of the system at scale, and using this failure dataset to improve the robot perception over time.
    • Preferred Majors: Aerospace & Mechanical Engineering, Astronautical Engineering, Computer Science, Electrical & Computer Engineering
    • Prerequisites: Comfort with coding and math; Past experience (through research work or coursework) in either machine learning, control theory, or robotics is a plus; Past experience of working with real robots is a plus

• Faculty/PI: Stephen Tu
• Website: https://stephentu.github.io
• Research Overview: My lab is interested in the interplay between statistical learning, dynamical systems, and control theory. Our goal is to rigorously understand the effects of integrating machine learning components into complex autonomous systems. While learning allows feedback systems to react to diverse sources of inputs (vision, speech, natural language, etc.), enabling many new and exciting capabilities, it also introduces complexity into the system which must be reasoned about in novel ways. At our core, we develop mathematical tools and practical algorithms to characterize the closed-loop behavior of dynamical systems with uncertainty.
• Summer Project:
  • Neural Galerkin Methods for Solving PDEs
    
    • Description: Numerical solutions to partial differential equations (PDEs) allow one to answer many important questions about the dynamics of closed-loop systems, including reachability, stability, and safety properties. Since PDEs are also important in various aspects of science, engineering, and mathematics, many numerical methods have been invented to solve them. One such method is the classic Galerkin method, which chooses a fixed basis and enforces the PDE condition on the subspace spanned by the basis. A typical choice for this fixed basis are localized functions gridding the domain of the PDE; this gives rise to finite-element algorithms for solving PDEs. However, as is well-known, these methods suffer from the curse of dimensionality. Can we overcome this in certain cases by using learnable basis functions, much as we do in machine learning? One idea which has come up recently in the literature is to use random basis elements; this is very similar to using random Fourier features from the machine learning literature. However, because the Galerkin method is based on linear algebraic arguments, the random basis is fixed and not trained to adapt to the data. The goal of this project is to design and implement algorithms to jointly learn both the features and the Galerkin weights. The selected candidate will first come up to speed on the background material, and then propose and evaluate new algorithms for numerical solutions to classic benchmark PDEs. Depending on the efficacy of the proposed algorithms, the project can then progress in various directions based on the candidate’s interest: either the candidate works on scaling up their solution to much higher dimensional problems, or the candidate embarks on a mathematical analysis to understand the theoretical properties of their proposed algorithm.
    • Student Responsibilities: Computational work and mathematical derivations.
    • Preferred Majors: Aerospace & Mechanical Engineering, Computer Science, Electrical Engineering, Mathematics, Physics, Statistics
    • Preferred Skillsets: Strong foundations in linear algebra, calculus, and probability theory. Background knowledge in machine learning and PDEs. Programming proficiency in Python and its scientific computing libraries (numpy, scipy, etc).

• Faculty/PI: Massoud Pedam
• Website: https://sportlab.usc.edu
• Research Overview:The System Power Optimization and Regulation Technology (SPORT) Lab focuses on research, development, and educational efforts on power-awareness in VLSI circuits and systems. Of particular interest are low-power electronics including low power devices, circuits, and architectures, computer-aided design methodologies and techniques for power-efficient realization of digital CMOS circuits, on-chip and off-chip bus encoding for low power, battery-powered mobile computing and communication, dynamic power management, and power-aware system-on-chip designs. The cross-disciplinary research spans the full scope of low power design technologies, including modeling and analysis of power dissipation sources, power conversion and regulation issues, IC design, system integration and PCB layout, system validation and online performance monitoring.
• Summer Projects
  • Expedition: DISCoVER
    • Project Description: uperconductor electronics (SCEs) are a transformative technology that pushes the limits of computation beyond CMOS. A family of SCE logics, single flux quantum (SFQ) logic, is based on Josephson junction (JJ). Unlike other technologies, SFQ is a nearly seamless replacement for classic CMOS, with logic families mirroring many CMOS capabilities. These platforms can be programmed to be transparent to developers, ensuring broad adoption. SCEs offer disruptive performance and power benefits, with switching speeds in the hundreds of GHz and minimal energy dissipation. The proposed DISCoVER Expedition aims to capitalize on this trajectory, exploring superconductivity for beyond-exascale computing. The SuperSoCC system, featuring a 32-bit superconducting CPU (SuperCPU) integrated with various components like a superconductive neural-network accelerator (SuperNN), aims for a significant performance-energy efficiency gain compared to CMOS. The expedition involves contributions from multiple institutions focusing on materials, devices, circuits, architecture, and integration to validate the SuperSoCC platform for beyond-exascale computing.
    • Student Responsibilities: Based on the student's interests and capabilities, they will work with one of the groups involved in architecture, circuit, or device development on CPU, NN, or Ising machine core. Students will work with Postdoc and Ph.D. students in the lab, helping them.
      Preferred Majors:  Computer Science, Electrical & Computer Engineering
    • Prerequisites and Preferred Skillsets:  Python, CPP programming skills; Familiar with hardware architecture; Good mathematics and physics background; Interested in novel technologies

• Faculty/PI: Andreas Molisch
• Website: wides.usc.edu
• Research Overview: The lab performs research on wireless communications, ranging from propagation channel measurement and modeling, to machine learning for video caching systems.
• Summer Projects:
  • Analysis of wireless communication channels under realistic conditions for 5G and beyond communication systems
    • Project Description: This project centers on measurement of propagation channels for 6G wireless systems. This will include measurements between drones and ground, measurements in the sub-THz frequency range, and measurements for distributed (cell-free) massive MIMO systems.
    • Student Responsibilities: Candidates will be working on configuring RF equipment, performing measurement campaigns, and/or processing measurement data for Channel Models.
    • Preferred Majors: Electrical & Computer Engineering
    • Preferred Skillsets: Candidates should preferably:
      • Be proactive
      • Be willing to work on teams
      • Have knowledge of Statistics and/or DSP is a plus
      • Have good programming skills in any of the following languages: MATLAB, LabVIEW, C++, P

• Faculty/PI: Shinyi Wu
• Summer Project
  • Data Integration and Model Development for Problematic Cannabis Use
    • Project Description: Our research focuses on understanding the patterns of Problematic Cannabis Use (PCU) among young adults in the dynamic environment of Los Angeles County, one of the nation's largest markets for liberalized cannabis use. We aim to develop a mathematical model by integrating the longitudinal behavioral survey data, theory-driven conceptual frameworks, and inputs from policy and community stakeholders. Our objective is to respond to national calls for advanced analytical techniques to identify risk and protective factors of PCU, examining how various elements influence the progression of cannabis use. These insights are crucial for shaping future cannabis policies and treatment practices.
    • Student Responsibilities:  Participants will gain hands-on experience in processing real-world longitudinal data and developing simulation models grounded in their theoretical understanding. This opportunity offers engagement in interdisciplinary research at the intersection of public policy, community health, and engineering approaches.
    • Preferred Majors:  Industrial & Systems Engineering
    • Prerequisites and Preferred Skillsets: Strong foundation in mathematics and comfort with learning mathematical modeling techniques. Proficiency in Python programming languages is preferred.

• Faculty/PI: Meisam Razaviyayn
• Website: https://sites.usc.edu/razaviyayn/group/
• Research Overview: Designing and studying of efficient large scale data analysis algorithms for machine learning tools and models. Led by Professor Meisam Razaviyayn
• Summer Projects:
  • Scalable fair and private learning in the presence of distribution shift
    • Project Description: Machine learning models are increasingly used in high-stakes decision-making systems. In such applications, a major concern is that these models sometimes discriminate against certain demographic groups such as individuals with a certain race, gender, or age. Another major concern in these applications is the violation of the privacy of users. While fair learning algorithms have been developed to mitigate discrimination issues, these algorithms can still leak sensitive information, such as individuals' health or financial records. On top of these challenges, there can be changes in data distribution when we move from training data to test data. In this project, we aim at developing scalable algorithms for tackling these challenges. This project is a part of a bigger collaborative project with the involvement of students from USC and researchers from industry (Meta, Amazon, Google Research).
    • Student Responsibilities: TBD
    • Preferred Majors: Computer Science
    • Prerequisites: Understanding the basics concepts in machine learning, optimization
    • Knowledge of Python and PyTorch/TensorFlow

• Faculty/PI: Mayank Kejriwal
• Website: https://viterbi.usc.edu/directory/faculty/Kejriwal/Mayank
• Research Overview: Kejriwal's research focuses on knowledge graphs (KG), an exciting area of Artificial Intelligence and data analytics research that has found widespread applications in industry (including in e-commerce giants like Amazon, and search providers like Google), academia (health informatics and social sciences) and for social causes (fighting human trafficking and mobilizing resources in the aftermath of crises). Simply put, knowledge graphs are a means to getting a machine to retain and 'understand' knowledge, rather than just raw data. Today, we live in an era when the World Wide Web provides us with an ever-expanding repository of data, yet we are still far from building machines that can process and understand this data in the way that domain experts (or in many cases, ordinary humans) can. Dr. Kejriwal has built and deployed knowledge graph-based systems that have been used by law enforcement and other subject matter experts to fight human trafficking, in addition to tackling other important domains like e-commerce and natural disaster response. Dr. Kejriwal draws on interdisciplinary research inspired by multiple communities in AI and data science, including human-computer interaction, social media, computational social science and Web sciences.

• Faculty/PI: Yolanda Gil
• Website: https://knowledgecaptureanddiscovery.github.io/
• Research Overview: We develop Artificial Intelligence (AI) approaches that use knowledge to accelerate and innovate scientific discovery processes that are unnecessarily carried out manually and inefficiently today. We work on a variety of AI research areas, such as semantic workflows, human-guided machine learning, interdisciplinary model integration, knowledge networks, controlled crowdsourcing of metadata, and automated hypothesis-driven discovery. A key theme in our projects is the use of AI technologies for different aspects of data science processes in order to make them more efficient. We collaborate with scientists in diverse areas including Earth sciences, neuroscience, genomics and proteomics, agriculture, and economics.
• Summer Projects:
  • Automating data science
    • Description: When scientists analyze data they do a lot of tasks that could be automated. We are developing intelligent systems that help scientists and data analysts to make sense of their data faster but also to surprise them with new interesting methods automatically synthesized through AI.  These AI systems will also help people who have no expertise in data analysis but have a lot of data, by automating data analysis for them and offering explanations of the findings.  We work with neuroscience, biomedicine, climate, and social sciences data.
    • Student Responsibilities: Students will apply existing AI systems to new types of data and domains, create user interfaces to explain how the AI system works, and generate visualizations that best illustrate new discoveries.
    • Preferred Majors: Computer Science,Electrical & Computer Engineering,Industrial & Systems Engineering
    • Prerequisites: Programming skills in Python; some experience using data for projects; and curiosity and interest in AI.