Summer Research Projects

Prof. Dong Song

Faculty Email: dsong@usc.edu

Name of Research Lab/Center: USC Neural Modeling and Interface Lab

Department: Biomedical Engineering

Research Lab Website: https://slab.usc.edu/

Projects:

1. Record and modeling neural signals for building memory prosthesis: We are the USC Neural Modeling and Interface Lab (https://slab.usc.edu) led by Dr. Dong Song. Our mission is to develop brain-like devices that can mimic and restore cognitive functions. To pursue this goal, we use a combined experimental and computational strategy to (1) understand how nervous systems such as the hippocampus perform higher-order cognitive functions, (2) develop next-generation modeling and neural interface methodologies to investigate brain functions during naturalistic behaviors, and (3) build cortical prostheses that can restore cognitive functions lost in diseases or injuries. We invite undergraduate students with science or engineering background to participate in these research projects. Each undergraduate researcher is expected to spend 8 to 10 hours per week (20 - 25 hours per week for the summer semester) for each semester on the project. Each student will have plenty of chances to participant in different aspect of the neuroscience and neural engineering projects described above under the supervision of Dr. Song and graduate students in the lab. Undergraduate researchers will be trained in the following skills:

Skill 1. in vivo Recording from Behaving Animals
Starting from preparing animals for MEA implantations, undergraduate researchers will assist graduate students to get the animal familiar with different experimental environments and train the animals to perform certain tasks such as free exploring and the DNMS task. Once the animal is ready for electrode implantation, undergraduate researchers will help to prepare chemicals needed for the implantation and sterilize the surgical area and the surgical instruments. During the surgery, undergraduate researchers will observe the entire implantation procedure and help to keep notes during the surgery. They will also learn detailed surgical techniques from graduate students. During the recovery period, undergraduate researchers will help to monitor the animal and report any abnormal condition to graduate students. After recovery, undergraduate researchers will assist graduate students to collect neural and behavioral data from implanted animals.

Skill 2. Histology of Brain Tissue
Performing histological study to brain tissue collected from implanted animals is a crucial step in neuroscience studies. This step identifies the origin of neural signals and is a necessary procedure to verify immune responses to the implant. First, undergraduate researchers will learn how to slice brain tissue and stain them with different dyes. Second, they will be trained to conduct histological works independently. Finally, they will learn to evaluate brain slices under microscope and keep documentations for each implanted animal.

Skill 3. Neural Data Analyze
Undergraduate researchers will learn to analyze neural data using specific software under from graduate students. They will assist in pre-processing neural data such as removal of artifact, counting units, and visualization of neural data using Matlab. Students with strong mathematic background and programming skills will also be able to participant in the development of computational models.

Prof. Birendra Jha

Faculty Email: bjha@usc.edu

Name of Research Lab/Center: Geosystems Engineering and Multiphysics Lab

Department: MFD Chemical Engineering and Materials Science

Research Lab Website: https://gemlab.usc.edu

Projects:

1. Fluid flow and mechanical deformation in porous media: experiments and simulations: Flow of water, oil and gas through geologic porous media (rocks and soil) is important for extraction of energy from underground, disposal of chemical waste, carbon sequestration to mitigate global warming, and for improving groundwater quality. Fluid flow through a geological, biological, or synthetic porous medium often leads to stress and strain in the medium, which affects the structure and properties of the medium, for example, its capacity to store and transmit fluids. This creates a fascinating world of beautiful patterns that emerge from physical interactions between the fluids and the porous medium. In this project, you will learn to create and observe those patterns in controlled laboratory experiments and computer simulation models. You can choose to sharpen your skills in differential equations, linear algebra, python/matlab programming, or fluid injection experiments. You will receive guidance from other students and researchers working in our lab. For experimental work, you will prepare samples for experiments (rock samples or acrylic plates), prepare fluids (water, glycerol, CO2), setup injection pump and coreholder, run experiments, collect data, make plots, and analyze. For computer simulation work, you will learn different components of a typical finite element simulator (initial and boundary conditions, material properties, mesh), prepare simulation input files, submit them to a computing cluster, collect output files, make plots, and analyze. Depending on your interests and progress, you may write computer codes or create new experiments to improve efficiency or understanding of the process. You will learn to work with a team of researchers, ask insightful questions, make hypotheses, and gather information necessary to test those hypotheses.

Prof. Jesse Thomason

Department: Computer Science

Research Lab Website: http://glamor.rocks/

Projects:

1. Grounding Language with Robots: Our lab is broadly interested in connecting language to agent perception and action, and lifelong learning through interaction. We work primarily at 3 levels of connecting language to the world: language and perception, understanding how words correspond to images, sounds, and feelings; language and actions, understanding how language can be used to instruct an agent about its goals and to communicate constraints; and language as a social mechanism for cooperating and learning. Students will work with PhD students on research-focused projects that involve connecting language to perception, actions, and social behavior. Students will work with large-scale language models, deep learning, simulators, and possibly physical robots, pending their and PhD students' interests by summer 2023.

Prof. Meisam Razaviyayn

Faculty Email: razaviya@usc.edu

Name of Research Lab/Center: Optimization for Data Driven Science

Department: ISE

Research Lab Website: https://sites.usc.edu/razaviyayn/group/

Projects:

1. Scalable fair and private learning in the presence of distribution shift: 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).

Prof. Somil Bansal

Faculty Email: somilban@usc.edu

Name of Research Lab/Center: Safe and Intelligent Autonomy Lab

Department: ECE

Research Lab Website: https://smlbansal.github.io/sia-lab/

Projects:

1. Vision-based navigation in new environments: 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/.

2. Safe assurances for learning and vision-driven robotic systems: 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.

Prof. Bhaskar Krishnamachari

Faculty Email: bkrishna@usc.edu

Name of Research Lab/Center: Autonomous Networks Research Group

Department: Electrical and Computer Engineering, and Computer Science

Research Lab Website: https://anrg.usc.edu

Projects:

1. Networked Distributed Systems and Applied Machine Learning: The Autonomous Networks Research Group directed by Prof. Bhaskar Krishnamachari, seeks bright undergraduate students with backgrounds in electrical engineering, computer science and mathematics for research into a broad range of topics including Internet of Things, robotic networks, connected vehicles, blockchain technology and applied machine learning. Projects will involve a mix of mathematics, simulation and testbed experiments, tailored to student background and interest, and provide a solid exposure to graduate-level research. Previous summer interns with this group have gone on to Ph.D. programs at top places including UC Berkeley, Columbia, MIT, Princeton, UCLA, USC, UT Austin, Stanford, U. Michigan, and UIUC.

Prof. Sifat Muin

Faculty Email: muin@usc.edu

Name of Research Lab/Center: Main Research Group

Department: Civil and Environmental Engineering

Research Lab Website: https://www.sifatmuin.com/

Projects:

1. Printed strain sensors

Prof. Ajitesh Srivastava

Faculty Email: ajiteshs@usc.edu

Name of Research Lab/Center: Srivastava Group

Department: Ming Hsieh Department of Electrical and Computer Engineering

Research Lab Website: www.ajitesh-srivastava.com

Projects:

1. Data Science for Health: The project will involve applications of machine learning, mathematical modeling, and data engineering to address problems in health, including epidemics and mental disorders. This may include time-series, networks, and ensemble techniques. Student duties include Literature review, coding, preparing presentations, weekly meetings.

Prof. Quan Nguyen

Faculty Email: quann@usc.edu

Name of Research Lab/Center: Dynamic Robotics and Control Lab

Department: Aerospace and Mechanical Engineering

Research Lab Website: https://sites.usc.edu/quann/

Prof. Ketan Savla

Faculty Email: ksavla@usc.edu

Name of Research Lab/Center: Savla Research Group

Department: Civil and Environmental Engineering

Research Lab Website: https://viterbi-web.usc.edu/~ksavla/

Prof. Luis Garcia

Faculty Email: lgarcia@isi.edu

Name of Research Lab/Center: STEEL: Security Research Lab

Department: Information Sciences Institute

Research Lab Website: https://steel.isi.edu/

Projects:

1. Cyber-physical Fuzzing Framework: Fuzzing is a standard automated technique for testing software for unwanted or unexpected behavior against different software inputs. When we put software on cyber-physical systems, i.e., systems that interact with the physical world through sensors and actuators, fuzzing the software is not sufficient for triggering unwanted behaviors. The sense-to-actuate pipeline is vulnerable to cyber-physical side-channels, such as sensor spoofing or false data injection. Conversely, we cannot just subject the physical systems to various signal generators as the systems would be subject to damage and could be costly. If we opt for simulation, modeling these behaviors would require capturing the sensor and actuator physical dynamics with high fidelity. The models would also need to understand how the sensor and actuator signals interact with the software. Students will study existing techniques to capture sensor and actuator dynamics in simulation and design and implement a prototype on an existing cyber-physical system testbed that can incorporate these models. Students will work with interfacing game-engine-based drone simulators with cybersecurity program analysis tools. This project is part of ongoing efforts and, thus, by the time the program starts, the student will work alongside PhD students to progress the state of the project.

2. Decoding how humans encode memory: Recent advancements in closed-loop deep brain stimulation (DBS) have enabled more intelligent autonomy for therapeutic intervention across a wide range of neurologic and psychiatric disorders. The predominant approach relies on control-theoretic approximations of the brain’s complex functional relationships with the external environment–-in particular, a mapping between targeted stimulation and naturalistic responses of different regions of the brain. However, existing approaches fail to capture the environmental context of neuronal biomarkers. Students will explore mechanisms to reason about complex sensor data from our recording platform, NeurIoT, that leverages a comprehensive set of Internet-of-Things (IoT) sensors to capture the human experience and environmental context, i.e., a subset of human sensory channels, in order to estimate the state of the human brain and provide the foundation for smarter, context-dependent DBS.

Prof. Ramesh Govindan

Faculty Email: ramesh@usc.edu

Name of Research Lab/Center: Networked Systems Laboratory

Department: Computer Science

Research Lab Website: https://nsl.usc.edu

Prof. Muhao Chen

Faculty Email: muhaoche@usc.edu

Name of Research Lab/Center: Language Understanding and Knowledge Acquisition Lab

Department: Computer Science

Research Lab Website: https://luka-group.github.io/

Projects:

1. Indirect Supervision for Knowledge Acquisition: Knowledge acquisition tasks (e.g., relation extraction, entity and event extraction and typing, consolidation) face challenges including extreme label spaces, insufficient annotations and out-of-distribution prediction. To this end, we study the method for leveraging indirect supervision signals from auxiliary tasks (e.g., natural language inference, abstractive summarization, etc.) to foster robust and generalizable inference for (open-domain) knowledge acquisition or information extraction. In the same context, study the method for generating semantically rich label representations based on either gloss knowledge or structural knowledge from a well-populated lexical knowledge base, in order to better support learning with limited labels.

Prof. Viktor Prasanna

Faculty Email: prasanna@usc.edu

Name of Research Lab/Center: Data Science Lab FPGA Lab

Department: ECE

Research Lab Website: 1. dslab.usc.edu   2. fpga.usc.edu

Projects:

1. Graph Machine Learning: Defining Graph ML models for applications. Predictive modeling. Optimization. Evaluation.

2. FPGA Acceleration of ML: Design and analysis of accelerators for ML tasks.

Prof. Hangbo Zhao

Faculty Email: hangbozh@usc.edu

Name of Research Lab/Center: Zhao Research Group

Department: Aerospace and Mechanical Engineering

Research Lab Website: https://sites.usc.edu/zhaogroup/

Projects:

1. Stretchable strain sensors for 3D reconstructions of soft robots: We develop stretchable sensors to measure large deformations of soft robots, then use the measured data to build 3D digital models of the soft robots. Student duties include: 1. Assist with strain sensor fabrication and testing
2. Assist with fabrication and actuation of soft robots
3. Assist with building 3D digital models of soft robots.

Prof. Peter Beerel

Faculty Email: pabeerel@usc.edu

Name of Research Lab/Center: Energy-Efficient Secue Sustainable Computing Group (EESSC)

Department: Electrical and Computing Engineering

Research Lab Website: https://sites.usc.edu/eessc/

Prof. Jelena Mirkovic

Faculty Email: mirkovic@isi.edu

Name of Research Lab/Center: STEEL

Department: USC Information Sciences Institute

Research Lab Website: steel.isi.edu

Prof. Mayank Kejriwal

Faculty Email: kejriwal@isi.edu

Name of Research Lab/Center: Artificial Intelligence and Complex Systems

Department: Information Sciences Institute

Research Lab Website: New website is under construction and will be available in early January. At present please use this: https://viterbi.usc.edu/directory/faculty/Kejriwal/Mayank

Prof. Mengjie Yu

Faculty Email: mengjiey@usc.edu

Name of Research Lab/Center: The Yu lab: integrated nonlinear and quantum photonics lab

Department: ECE

Research Lab Website: https://sites.usc.edu/mjlab/

1. Integrate photonics

Prof. Dominique Duncan

Faculty Email: duncand@usc.edu

Name of Research Lab/Center: Duncan Lab

Department: Neuroimaging and Informatics

Research Lab Website: https://sites.usc.edu/duncanlab/

Prof. Paul Bogdan

Faculty Email: pbogdan@usc.edu

Name of Research Lab/Center: Cyber-Physical Systems

Department: ECE

Research Lab Website: https://cps.usc.edu/studs.html

Prof. BO Jin

Faculty Email: bochengj@usc.edu

Name of Research Lab/Center: Advanced Composites Design Lab

Department: Aerospace and Mechanical Engineering

Research Lab Website: composites.usc.edu

Prof. Jason Kutch

Faculty Email: kutch@usc.edu

Name of Research Lab/Center: Applied Movement and Pain Laboratory

Department: Biokinesiology and Physical Therapy

Research Lab Website: ampl.usc.edu

Projects:

Looking for students from these departments: Aerospace & Mechanical Engineering,Biomedical Engineering,Computer Science,Electrical & Computer Engineering

Prof. Vatsal Sharan

Faculty Email: vsharan@usc.edu

Name of Research Lab/Center: ML Theory Group

Department: Computer Science

Research Lab Website: https://vatsalsharan.github.io/

Projects:

1. Machine Learning Mysteries: he student 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.

Familiarity with probability, linear algebra, calculus, and analysis of algorithms. Some basic understanding of machine learning would be very helpful for certain projects.

Prof. Wade Hsu

Faculty Email: cwhsu@usc.edu

Name of Research Lab/Center: Photonics in Complex Systems Group

Department: Electrical and Computer Engineering

Research Lab Website: https://sites.usc.edu/hsugroup/

Projects:

1. Non-invasive 3D Imaging inside Biological Tissue: In this project we investigate the design of new ray tracing/ ray launching algorithms for efficient prediction of wireless propagation in 5G/6G scenarios. One of the aspects we consider is the application of cutting-edge computer science methods for the acceleration of ray tracing. Another is the design of ray launching simulations that can handle movement of not only the receiver (where movement can be incorporated easily by traditional approaches), but also transmitter and environmental objects.

Two students will be working on particular on the topic of ray visibility and object edge detection, making use of computer science approaches (these positions are thus best suited for CS students). They will research, implement, and test these algorithms in Python or a similar language. Another student (ECE preferred) will be performing ray tracing simulations with a commercial ray tracer, designing the simulation setups to accommodate movement of transmitter and objects in an efficient way. Experience with ray tracing algorithms, edge detection, etc., are ideal. For the two CS-oriented positions, extensive programming experience/skill is required. For the ECE-oriented position, interest in (and preferable experience in) wireless propagation (electromagnetics) is desirable.

2. Fast multi-source nanophotonic simulations: Maxwell's equations describe phenomena over the full electromagnetic spectrum from visible light to radio waves. Numerous problems, such as optical computing, metasurface design, inverse-scattering imaging, and stealth aircraft design, require computing the scattered wave given a very large number of distinct incident waves. However, existing Maxwell solvers scale poorly--either in computing time or in memory--with the number of input states of interest. We are developing highly efficient multi-source solvers for Maxwell's equations. The student will take part in code development, testing, and preparing tutorials for our MESTI solver (https://github.com/complexphoton/MESTI.m).

3. Ab-initio theory of lasers near exceptional points: "Exceptional points" are unique states where two optical modes coalesce into one mode. Lasers operating near exceptional points exhibit enhanced sensitivity to perturbations, making them promising as ultra-sensitive sensing devices such as gyroscopes. However, existing laser theories cannot describe lasers near EP, because such lasers have a non-stationary population inversion. We are developing an ab-initio laser theory that can describe such complex lasers. The student will take part in the theoretical and numerical modeling of the nonlinear dynamics and noise properties of such unique lasers.

4. Inverse design of high-speed spatial-light modulators: The student will take part in code development and performing simulations and the numerical inverse design of the high-speed SLM.

Prof. Andreas Molisch

Faculty Email: molisch@usc.edu

Name of Research Lab/Center: WiDeS

Department: Electrical and Computing Engineering

Research Lab Website: wides.usc.edu

Projects:

1. Ray tracing for wireless propagation research: In this project we investigate the design of new ray tracing/ ray launching algorithms for efficient prediction of wireless propagation in 5G/6G scenarios. One of the aspects we consider is the application of cutting-edge computer science methods for the acceleration of ray tracing. Another is the design of ray launching simulations that can handle movement of not only the receiver (where movement can be incorporated easily by traditional approaches), but also transmitter and environmental objects.

2. Wireless propagation channel measurements: In this project we will perform measurements of wireless propagation channels with various channel sounders. These can range from measurements based on existing WiFi signals to measurements in the higher frequency ranges.

Prof. Hossein Hashemi

Faculty Email: hosseinh@usc.edu

Name of Research Lab/Center: HHLab

Department: Electrical and Computing Engineering

Research Lab Website: https://hhlab.usc.edu/

Projects:

1. Millimeter-Wave Integrated Circuits: The student will design and simulate millimeter-wave integrated circuits.

2. Silicon Photonics Integrated Circuits: The project involves application of silicon photonics integrated circuits for lidar, optical computing, and other applications.

Prof. Pierluigi Nuzzo

Faculty Email: nuzzo@usc.edu

Name of Research Lab/Center: Cyber-Physical System Design (DesCyPhy) Lab and Center for Autonomy and Artificial Intelligence (CAAI)

Department: Electrical and Computer Engineering and Computer Science

Research Lab Website: https://descyphy.usc.edu/

Projects:

Looking for students from the following departments: Aerospace & Mechanical Engineering,Computer Science,Electrical & Computer Engineering.

Prof. Stefanos Nikolaidis

Faculty Email: nikolaid@usc.edu

Name of Research Lab/Center: Interactive and Collaborative Autonomous Robotic Systems

Department: Computer Science

Research Lab Website: http://icaros.usc.edu

Projects:

1. Human-Robot Collaboration in Shared Workspaces: We focus on enabling robots to assist people in collaborative tasks, such as assembling an IKEA bookcase: https://youtu.be/wDHQXBua4OI
This requires robot assistants to perceive parts in the environment, grasp the parts while avoiding obstacles and hand the parts safely to human teammates. Robots also need to observe the actions of their human counterparts, infer their intent and preemptively take actions to help them in the task, e.g., deliver a screw driver when the user is about to fasten a screw. The tight integration of perception, inference and action results in seamless collaboration between a robot assistant and a human user. The students will work on programming the robot behaviors in simulation environments and in the real world. They will develop, implement and improve algorithms for robot learning and planning. They will then deploy the algorithms to our robotic arms in the lab and test them in a variety of collaborative tasks. Required skill: Programming ability in Python. Required coursework: Probability Theory, Calculus.
Preferred (but not required) coursework: Algorithms, Machine Learning, Robotics.

Prof. Yasser Khan

Faculty Email: yasser.khan@usc.edu

Name of Research Lab/Center: Khan Lab

Department: Electrical and Computer Engineering

Research Lab Website: khan.usc.edu

Projects:

1. Optoelectronic sensing with ingestibles: We are developing a smart capsule that will go in the gut and measure chemical markers. In this project, you will be aiding the electronic prototyping as well data collection of the capsule.

2. Stretchable MRI receive coils: We will be developing a fabrication process for manufacturing MRI receive coils using liquid metal conductors. USC has one of the world's three 0.55T low-field MRI systems. This MRI receive coil will be the first-ever flexible MRI received coil designed for a low-field MRI system. Fabrication and characterization of liquid metal electronics.

3. Printing and characterization of organic electrochemical transistors: We are developing next-generation printed sensors and circuits with organic electrochemical transistors. Students will fabricate and characterize printed OECT-based sensors and circuits.

Published on September 18th, 2017

Last updated on May 15th, 2023