About me
Personal Info
Welcome to my homepage! My name is Lingwei Zhu (朱 令緯). I am now a Project Researcher at the CDR lab, University of Tokyo working with Prof. Yukie Nagai. Prior to coming to Tokyo, I spent two wonderful years at the RLAI Lab, University of Alberta as a postdoc working with Prof. Martha White. I obtained my PhD with the Best Student Honor at Nara Institute of Science and Technology (NAIST) under the supervision of Prof. Takamitsu Matsubara.
My research interest lies in applications of artificial intelligence and machine learning for solving complex real-world problems. Especially reinforcement learning and unsupervised learning for autonomous control of large-scale systems (industrial processes and robotics) and automated biomedical systems (cancers, brain activity cognition).
My CV can be found here
Preprints and Working Papers
($\dagger$ indicates joint first authors)
Towards Physiologically Sensible Predictions via the Rule-based Reinforcement Learning Layer, [link]
Lingwei Zhu, Zheng Chen, Yukie Nagai, Jimeng Sun
Publications
Fat-to-Thin Policy Optimization: Offline RL with Sparse Policies, [link]
Lingwei Zhu$^{\dagger}$ , Han Wang $^{\dagger}$, Yukie Nagai
accepted by International Conference on Learning Representations (ICLR) 2025
$q$-Exponential Family for Policy Optimization, [link]
Lingwei Zhu$^{\dagger}$ , Haseeb Shah $^{\dagger}$ , Han Wang $^{\dagger}$ , Yukie Nagai, Martha White
accepted by International Conference on Learning Representations (ICLR) 2025
Offline Reinforcement Learning with Tsallis Regularization, [link]
Lingwei Zhu$^{\dagger}$ , Matthew Schlegel $^{\dagger}$ , Han Wang, Martha White
accepted by Transaction on Machine Learning Research (TMLR) 2024
Generalized Munchausen Reinforcement Learning using Tsallis KL Divergence, [link]
Lingwei Zhu, Zheng Chen, Matthew Schlegel, Martha White
accepted by Neural Information Processing Systems (NeurIPS) 2023
Cautious Policy Programming: Exploiting KL Regularization for Monotonic Policy Improvement in Reinforcement Learning, [link]
Lingwei Zhu, Takamitsu Matsubara
accepted by Machine Learning 2023
Cancer Subtyping by Improved Transcriptomic Features Using Vector Quantized Variational Autoencoder, [link]
Z Chen$^{\dagger}$, ZW Yang $^{\dagger}$, Lingwei Zhu$^{\dagger}$, G Shi, K Yue, Takashi Matsubara, S Kanaya, MD Altaf-Ul-Amin,
Computer Methods and Programs in Biomedicine (CMPB) 2023
Cyclic Policy Distillation: Sample-Efficient Sim-to-Real Reinforcement Learning with Domain Randomization, [link]
Y Kadokawa, Lingwei Zhu, Y Tsurumine and Takamitsu Matsubara,
Robotics and Autonomous Systems (RAS) 2023
Hierarchical Categorical Generative Modeling for Multi-omics Cancer Subtyping, [link]
Ziwei Yang$^{\dagger}$, Lingwei Zhu$^{\dagger}$, Chen Li, Zheng Chen, Naoki Ono, Md Altaf-Ul-Amin, Shigehiko Kanaya, IEEE International Conference on Bioinformatics and Biomedicine (BIBM) 2022
Automated Sleep Staging via Parallel Frequency-Cut Attention, [link]
Z Chen, ZW Yang, Lingwei Zhu, W Chen, T Tamura, N Ono, MD Altaf-Ul-Amin, S Kanaya, M Huang,
IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2022
Automated Cancer Subtyping via Vector Quantized Mutual Information Maximization, [link]
Z Chen$^{\dagger}$, Lingwei Zhu$^{\dagger}$, Ziwei Yang, Takashi Matsubara,
accepted by ECML-PKDD 2022
Multi-Tier Platform for Cognizing Massive Electroencephalogram, [link]
Z Chen$^{\dagger}$, Lingwei Zhu$^{\dagger}$, Ziwei Yang, Renyuan Zhang,
accepted by IJCAI 2022
Cancer Subtyping via Embedded Unsupervised Learning on Transcriptomics Data, [link]
Z Yang, Lingwei Zhu, Z Chen, M Huang, N Ono, MD Altaf-Ul-Amin, S Kanaya
accepted by EMBC 2022
Adaptive Spike-Like Representation of EEG Signals for Sleep Stages Scoring, [link]
Lingwei Zhu, K Odani, Z Yang, G Shi, Y Kan, Z Chen, R Zhang,
accepted by EMBC 2022
Alleviating Parameter-tuning Burden in Reinforcement Learning for Large-scale Process Control, [link]
Lingwei Zhu, G Takami, M Kawahara, H Kanokogi, T Matsubara,
Computers & Chemical Engineering 2022
Cautious Actor-Critic, [link]
Lingwei Zhu, T Kitamura, T Matsubara,
Asian Conference on Machine Learning 2021
Geometric Value Iteration: Dynamic Error-Aware KL Regularization for Reinforcement Learning, [link]
T Kitamura, Lingwei Zhu, T Matsubara,
Asian Conference on Machine Learning 2021
Dynamic actor-advisor programming for scalable safe reinforcement learning, [link]
Lingwei Zhu, Y Cui, T Matsubara,
IEEE ICRA 2020
Scalable reinforcement learning for plant-wide control of vinyl acetate monomer process, [link]
Lingwei Zhu, Y Cui, G Takami, H Kanokogi, T Matsubara,
Control Engineering Practice 2020
Factorial kernel dynamic policy programming for vinyl acetate monomer plant model control, [link]
Y Cui$^{\dagger}$, Lingwei Zhu$^{\dagger}$, M Fujisaki, H Kanokogi, T Matsubara,
IEEE CASE 2018
Professional Activities
Program Commitee Member (Reviewer)
JMLR, IEEE-TNNLS, TMLR, NeurIPS, ICLR, AAAI, IJCAI, ECML, RAL, ICRA, IROS