Yifeng Xiao

yifengx-at-usc.edu

I am a Ph.D. student studying Electrical Engineering at the University of Southern California. I am advised by Prof. Pierluigi Nuzzo of the DesCyPhy Lab. My research focuses on the design and verification of cyber-physical systems, utilizing contract theory for system abstraction, compositional verification, and AI-based methodologies for system design.

I received my B.E. in Microelectronic Science and Engineering from Fudan University. I was fortunate to work with Prof. Jianli Chen and Prof. Bei Yu on graph and machine learning-based applications in electronic design automation.

I’m looking for self-motivated MS students at USC to work on research projects. If you are interested in machine learning-based system design, please feel free to contact me.

Education

University of Southern California

Los Angeles, CA, USA

Ph.D. in Electrical Engineering

Advisor: Pierluigi Nuzzo

January 2021 - Present

Fudan University

Shanghai, China

B.E. in Microelectronic Science and Engineering

Advisor: Jianli Chen

August 2016 – June 2020

University of Sydney

Sydney, Australia

Exchange Student in Information & Computer Engineering

February 2019 – June 2019

Projects

Formalisms for Cyber-Physical System Specification, Analysis, and Design

DesCyPhy Lab

The goal of the project is to investigate modeling paradigms, algorithms, design methodologies, design flows, and supporting tools to enable compositional and hierarchical design of complex cyber-physical systems, with emphasis on systems enabled by artificial intelligence. We show that a contract-based approach can provide a formal foundation for system design methodologies encompassing both horizontal and vertical integration steps.

March 2021 - Present

Self-Driving Vehicle Testbed

AutoDrive Lab

The goal of this project is to build simulation-based and experimental testbeds to emulate realistic scenarios for self-driving vehicles and test the effectiveness of different driving algorithms. The testbeds will target traffic intersections and will include, for example, scaled-down autonomous cars, programmable traffic light sequencers to emulate the traffic and pedestrian signals, and robots to emulate pedestrian traffic.

September 2021 - Present