Sihong He

AI-enhanced Cyber-Physical Systems

prof_pic.png
sihong.he@uconn.edu

I am on the job market! Thank you for providing any recurring information!!!

Hi, my name is Sihong He, currently a 5th-year Ph.D. candidate at the Department of Computer Science and Engineering, University of Connecticut, working with Dr. Fei Miao. Before my Ph.D. journey at UConn, I received my M.S. degree in Statistics at UC Irvine in 2019 and my B.E. degree in Financial Mathematics in 2017 at SUSTC (南方科技大学) which is newly established in 2011 at Shenzhen, China. (click to read a short story about our university). I enjoy trying new things, learning new knowledge, and exploring the frontiers of research (which is also a reason why I studied in three different departments in the past 10 years :blush:).

My research interests include multi-agent reinforcement learning, robust and fair reinforcement learning, cyber-physical systems, and human-centered computing. My research goal is to lay the foundations for the collective intelligence of interconnected CPS, including ensuring efficiency, robustness, safety, and security for applications areas such as intelligent transportation systems, connected autonomous vehicles, autonomous delivery systems, and power networks. My research vision is to leverage AI and CPS a force for social good to elevate the quality of life, enhance societal well-being, and create a better future for people.

What did I do in my research?

(1) Construct theoretical frameworks and practical algorithms of learning-based methods (multi-agent reinforcement learning, deep learning) for robust and fair interconnected CPS (connected autonomous vehicles, mobility-on-demand systems, Smart grid, etc).

(2) Develop the theory and application of data-driven optimization methods for efficient mobile CPS (autonomous mobility-on-demand systems, autonomous delivery systems, etc).

(3) Ensure the safety and security of interconnected CPS through integrated learning and control.

!!Open to collaboration!! Please contact me if you are interested in Reinforcement Learning, Optimization, CPS, Federated Learning, GNN, and LLM. Let's do some interesting research together.

So far, my research has focused on robust multi-agent reinforcement learning for robust interconnected CPS, data-driven robust optimization for efficient mobile CPS, and on the security and safety of CPS. In addition to system modeling, theoretical analysis, and algorithm design, my work involves experimental validation of real-world data. My work has been published in prestigious journals and conferences including International Conference on Intelligent Robots and Systems (IROS), International Conference on Robotics and Automation (ICRA), IEEE Transactions on Intelligent Transportation Systems (TITS), ACM Transactions on Cyber-Physical Systems (TCPS), IEEE Transactions on Mobile Computing (TMC) and Transactions on Machine Learning Research (TMLR).

news

Oct 17, 2023 Our Journal paper “FairMove: A Data-Driven Vehicle Displacement System for Jointly Optimizing Profit Efficiency and Fairness of Electric For-Hire Vehicles” is accepted to IEEE Transactions on Mobile Computing. Congratulations and Many thanks to Guang!
Sep 8, 2023 It’s my honor to receive IROS Travel Grants. See you in Detroit.
Sep 1, 2023 I passed my proposal defense and am formaly a Ph.D. candidate now. Thanks to my committee members and all audience attending my oral presentation. 🥰
Jun 23, 2023 Two papers have been accepted to IROS2023! :sparkles: Check our papers here: “A robust and constrained multi-agent reinforcement learning framework for electric vehicle AMoD systems” and “Robust electric vehicle balancing of autonomous mobility-on-demand system: A multi-agent reinforcement learning approach”. Many thanks to my collaborators, Shuo, Shaofeng and Yue. See you in Detroit!
May 25, 2023 A journal paper “Robust Multi-Agent Reinforcement Learning with State Uncertainty” is accepted to Transactions on Machine Learning Research! :smile: We provide the first attempt at the theoretical and empirical analysis of robust MARL problem with state uncertainty in this paper.
Mar 8, 2023 Two papers are accepted to AI for Agent-Based Modelling (AI4ABM) Workshop at the International Conference on Learning Representations (ICLR) 2023 :smile:!
Jan 28, 2023 A conference paper is accepted to ICRA! :sparkles: :smile:
Jan 26, 2023 A journal paper “Data-Driven Distributionally Robust Electric Vehicle Balancing for Autonomous Mobility-on-Demand Systems Under Demand and Supply Uncertainties” is published on IEEE Transactions on Intelligent Transportation Systems!
Aug 28, 2019 Start my PhD journey!

selected publications

  1. A robust and constrained multi-agent reinforcement learning framework for electric vehicle AMoD systems
    Sihong He, Yue Wang, Shuo Han, and 2 more authors
    arXiv preprint arXiv:2209.08230, 2023
  2. Robust electric vehicle balancing of autonomous mobility-on-demand system: A multi-agent reinforcement learning approach
    Sihong He, Shuo Han, and Fei Miao
    arXiv preprint arXiv:2307.16228, 2023
  3. Robust Multi-Agent Reinforcement Learning with State Uncertainty
    Sihong He, Songyang Han, Sanbao Su, and 3 more authors
    Transactions on Machine Learning Research, 2023
  4. Data-Driven Distributionally Robust Electric Vehicle Balancing for Autonomous Mobility-on-Demand Systems Under Demand and Supply Uncertainties
    Sihong He, Zhili Zhang, Shuo Han, and 5 more authors
    IEEE Transactions on Intelligent Transportation Systems, 2023
  5. Data-driven distributionally robust electric vehicle balancing for mobility-on-demand systems under demand and supply uncertainties
    Sihong He, Lynn Pepin, Guang Wang, and 2 more authors
    In 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020