Laixi Shi
Postdoctoral Fellow, Caltech CMS
Email: laixis AT caltech DOT edu
Email: laixis AT caltech DOT edu
Bio
Hello, I am currently a postdoctoral fellow in the Dept. of Computing + Mathematical Sciences (CMS) at California Institute of Technology (Caltech), fortunately mentored by Prof. Adam Wierman and Prof. Eric Mazumdar. I graduated from the Dept. of Electrical and Computer Engineering from Carnegie Mellon University (CMU) on August 2023, where I am fortunate to be advised by Prof. Yuejie Chi. Before joining CMU, I received my bachelor's degree from Electronic Engineering, Tsinghua University in July 2018.
Research Interests
- Provable reinforcement learning: sample efficiency, robustness, and scalability: Thesis as a summary; online RL, offline RL(I and II), and robust RL(I and II).
- Practical reinforcement learning: offline RL(I and II), robust RL, and curriculum RL.
- (Nonconvex) optimization for signal processing: Multi-channel blind deconvolution, image stitching.
- Real-world data science solutions: indoor localization, strain measurement, hand gesture recognition.
News
[2023/10/17] A new website about practical robust RL has been released.
[2023/9/21] Two papers got accepted by NeurIPS 2023. Thanks for all wonderful collaborators. See you on Dec. 10-16.
[2023/08] Call for Papers at Conference on Parsimony and Learning (CPAL)!!! (serve as an Area Chair).
[2023/08/21] Computing, Data, and Society Postdoctoral Fellow at Caltech.
[2023/07/28] Release my thesis after Ph.D. defense.
Recent/Upcoming Events
[2024/02/18-23] 2024 Information Theory and Applications Workshop.
[2023/11/14-15] Rising Stars In Machine Learning, University of Maryland.
[2023/11/08] Invited speaker at WORDS 2023: Workshop in Operations Research and Data Science, Duke University.
[2023/10/26] Invited speaker at Safe Reinforcement Learning Online Seminar at 9am PDT (Talk Zoom link).
[2023/10/15-18] Invited speaker at 2023 INFORMS Annual Meeting.
Selected Preprints & Publications
Federated Offline Reinforcement Learning: Collaborative Single-Policy Coverage Suffices
Jiin Woo, Laixi Shi, Gauri Joshi, Yuejie Chi.
Preprint.
[Arxiv]
Settling the Sample Complexity of Model-Based Offline Reinforcement Learning
Gen Li, Laixi Shi, Yuxin Chen, Yuejie Chi, Yuting Wei.
Accepted by The Annals of Statistics.
[Paper]
Pessimistic Q-Learning for Offline Reinforcement Learning: Towards Optimal Sample Complexity
Laixi Shi, Gen Li, Yuting Wei, Yuxin Chen, Yuejie Chi.
International Conference on Machine Learning (ICML), 2022.
[Paper]
Breaking the Sample Complexity Barrier to Regret-Optimal Model-Free Reinforcement Learning
Gen Li, Laixi Shi, Yuxin Chen, Yuejie Chi
Information and Inference: A Journal of the IMA. A short version accepted as [NeurIPS Spotlight] Conference on Neural Information Processing Systems (NeurIPS), 2021
Offline Reinforcement Learning with On-Policy Q-Function Regularization
Laixi Shi, Robert Dadashi, Yuejie Chi, Pablo Samuel Castro, Matthieu Geist.
European Conference on Machine Learning (ECML), 2023.
[Arxiv]
Seeing is not Believing: Robust Reinforcement Learning against Spurious Correlation
Wenhao Ding*, Laixi Shi*, Yuejie Chi, Ding Zhao.
Conference on Neural Information Processing Systems (NeurIPS), 2023. A short version at the Second Workshop on Spurious Correlations, Invariance and Stability of International Conference on Machine Learning (ICML), 2023.
A Trajectory is Worth Three Sentences: Multimodal Transformer for Offline Reinforcement Learning
Yiqi Wang, Mengdi Xu, Laixi Shi, Yuejie Chi.
The Conference on Uncertainty in Artificial Intelligence (UAI), 2023.
[Paper]
Curriculum Reinforcement Learning using Optimal Transport via Gradual Domain Adaptation
Peide Huang, Mengdi Xu, Jiacheng Zhu, Laixi Shi, Fei Fang, Ding Zhao
Conference on Neural Information Processing Systems (NeurIPS), 2022
[Paper]
Manifold Gradient Descent Solves Multi-channel Sparse Blind Deconvolution Provably and Efficiently
Laixi Shi, Yuejie Chi
IEEE Transactions on Information Theory, vol. 67, no. 7, pp. 4784-4811, 2021. A short conference version has been accepted by International Conference on Acoustics, Speech, and Signal ProcessingInternational Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2020.
Latent Goal Allocation for Multi-Agent Goal-Conditioned Self-Supervised Imitation Learning
Laixi Shi*, Peide Huang*, Rui Chen*
NeurIPS Workshop on Bayesian Deep Learning, 2021 ( * = equal contribution)
[Paper]
Projects
Nov 1, 2016. Advisor: Huimin Ma, Tsinghua University, Beijing, China.
We designed a real-time virtual reality cubic interaction metaphor, arranged 5 LED panels into a cube to create a 3D object illusion, utilizing binocular vision to track user’s eyes and rendering the perspective of each screen to fit the user’s eyes positions.
[Video]
Jul 18, 2016. Advisor: Yuan Shen, Tsinghua University, Beijing, China
We developed a multiple robot collaborative localization and navigation system using a self-made mobile robot platform. The robots in our system are equipped with a wireless distributed localization system based on UWB(Ultra Wide Band), which enables them automatically to form a pre-defined formation with mutual localization information. (Image from my collaborator Wenhao Ding)
Jun 13, 2015, Tsinghua University, Beijing, China
We built a robot to play air-hockey game with human, which is a popular table game. This robot consists of a bird-view camera (capturing the images of moving ball) and an executive arm (a striking device with two degrees of freedom). (Image from my collaborator Wenhao Ding)
Awards
Honors / Awards / Fellowships
- Rising Stars In Machine Learning by University of Maryland (2023)
- Computing, Data, and Society Postdoctoral Fellow by Caltech (2023)
- ICASSP Rising Stars in Signal Processing (2023)
- UT Austin Rising Stars in Computational and Data Sciences (2023)
- UChicago Rising Stars in Data Science (2022)
- NeurIPS 2022 Top Reviewer (2022)
- Leo Finzi Memorial Fellowship (2022)
- Wei Shen and Xuehong Zhang Presidential Fellowship (2022)
- Liang Ji-Dian Graduate Fellowship (2021)
- Presidential Fellowship granted by Carnegie Mellon University (2018)
- Carnegie Institute of Technology Dean’s Fellowship (2018)
- Excellent Honors Graduate granted by Tsinghua University (2018)
- The First Prize in 35th Tsinghua University Academic Challenge Cup (2017)
- Technology Innovation Excellence Award granted by Tsinghua University (2015-2017)
- Enterprise Sponsored Scholarship granted by Tsinghua University (2017)
- National Scholarship granted by the government of China (2016)
- Qualcomm Scholarship granted by Tsinghua University (2016)
- Outstanding Project of Undergraduate Research Competition of Tsinghua University (2016)
- The First Prize in National Physics Contest for College Student (2015)
- The Silver Medal of Chinese Physics Olympiad (2014)
Travel Awards
- IAS WAM Mathematics in Machine Learning Travel Support (2022)
- National Science Foundation (NSF) Student Travel Grants for ICASSP (2020)
- Computing Research Association Grad Cohort for Women Travel Support (2020)
- The ACM International Workshop on Device-Free Human Sensing Travel Grant (2019)
- Women in Machine Learning Scholarship (2019)
- Women in Data Science and Mathematics Travel Support (2019)