Yixuan Wang

alt text

Hi, welcome to my personal page!

I'm a fifth-year Ph.D. student in the Computer Engineering at Northwestern University, supervised by Prof. Qi Zhu. I also work closely with Prof. Chao Huang from Unversity of Liverpool. My research interestes include machine learning and formal methods, with the consideration of safety, stability, robustness and correctness for cyber-physical systems.

Prior to Northwestern, I received my B.E. from Tsinghua University in 2018.

E-mail: yixuanwang2024 [AT] u [DOT] northwestern [DOT] edu

My research overview, Curriculum Vitae, Google Scholar, Github

Recent News

  • 2024.05 Our paper Boosting Long-Delayed Reinforcement Learning with Auxiliary Short-Delayed Task is accepted to ICML 2024.

  • 2024.04 I passed my Ph.D. defense exam! The slide presentation can be found here.

  • 2024.03 Our paper State-wise Safe Reinforcement Learning with Pixel Observations has been accepted to L4DC.

  • 2024.03 Our paper Empowering Autonomous Driving with Large Language Models: A Safety Perspective is accepted by ICLR 2024 Workshop on LLM Agents.

  • 2024.01 We have a few preprints addressing various problems including safe RL with image observations, trajectory prediction for autonomous vehicles, and large language model for autonomous driving.

  • 2023.12 Our paper REGLO: Provable Neural Network Repair for Global Robustness Properties has been accepted by AAAI 2024.

  • 2023.10 Our paper POLAR-Express: Efficient and Precise Formal Reachability Analysis of Neural-Network Controlled Systems is accepted to TCAD.

  • 2023.08 The co-authored book chapter Safety-Assured Design and Adaptation of Connected and Autonomous Vehicles is online.

  • 2023.06 Our paper Safety-Assured Speculative Planning with Adaptive Prediction is accepted to IROS 2023, see u in Detroit this October!

  • 2023.05 I will give a talk to the 6th IEEE International Workshop on Design Automation for Cyber-Physical Systems (DACPS), in July. The title of this talk is “Explainable safe reinforcement learning for safety-critical CPSs” which will mainly cover our recent ICML and ICCPS paper, where we developed a differentiable joint-learning bi-level optimization framework for safe RL with both deterministic and stochastic environment.

  • 2023.04 Our paper Enforcing Hard Constraints with Soft Barriers: Safe Reinforcement Learning in Unknown Stochastic Environments has been accepted to ICML 2023, see u in Hawaii this Summer!

  • 2023.03 I passed the Ph.D. Prospectus Exam!

  • 2023.01 Our paper Joint Differentiable Optimization and Verification for Certified Reinforcement Learning is accepted by ICCPS 2023.

Working Experience

  • Aurora Innovation, Mountain View, CA, USA, 2023.07 - 2023.09, Machine Learning Engineering Intern

  • Bosch USA, Sunnyvale, CA, USA, 2023.03 - 2023.06, Machine Learning Engineering Intern

  • Ford Motor Company, Dearborn, MI, USA, 2021.06 - 2021.09, Research Intern

  • Sensetime, Shenzhen, China 2019.04 - 2019.08, Machine Learning Engineering Intern

  • Mech-Mind Robotics, Beijing, China, 2017.12 - 2018.02, Software Development Engineering Intern

Services

  • Reviewer for AAAI, NeurIPS, ICML, ICLR, RAL, CDC, IV, IOT Journal, etc

  • ICCPS 2024 Poster/Demo PC member

  • ICCPS 2024 Artifact Evaluation PC member

Teaching

  • Teaching Assistant for ELEC_ENG 373, 473 Deep Reinforcement Learning from Scratch, Spring 2022.

Awards

  • Terminal Year Fellowship, Nothwestern University, 2023.

  • Best Paper Candidate, ICCAD 2020.

  • Northwestern Univerisity Ph.D. Fellowship, 2019-2020.

  • Academic Excellence Award(top 10%), Tsinghua University, in all the undergraduate years(2014-2018).

  • National Scholarship(top 2%), 2016.

  • Outstanding Undergraduate Students(top 2%), Tsinghua Univerisity, 2018.

  • Outstanding Undergraduate Students of Beijing(top 5%), 2018.