Ruipeng Zhang

Hi! I am a Ph.D. student in Computer Science at UC San Diego, advised by Prof. Sicun Gao.

I am broadly interested in reinforcement learning and robotics. Specifically, my work focuses on making reinforcement learning more sample efficient and enabling robots to learn useful behaviors from limited demonstrations. I was also an Applied Scientist Intern at Amazon’s Industrial Robotics Group in summer 2026.

Selected Publications

MoRR overview: robot manipulation stages, policy routing weights, and residual policy improvement architecture

ICRA 2027 · Under review MoRR: Mode-Aware RL via Residual Policy Improvement Routing

Ruipeng Zhang, Sarthak Kaingade, Zhenyu Liao, Baljeet Singh, Sicun Gao, Bryan Trinh

ICRA 2026 Learning Quadruped Walking from Seconds of Demonstration

Ruipeng Zhang, Hongzhan Yu, Ya-Chien Chang, Chenghao Li, Henrik I. Christensen, Sicun Gao

arXiv Website Code

BibTeX

@article{zhang2026learning, title={Learning Quadruped Walking from Seconds of Demonstration}, author={Zhang, Ruipeng and Yu, Hongzhan and Chang, Ya-Chien and Li, Chenghao and Christensen, Henrik I and Gao, Sicun}, journal={arXiv preprint arXiv:2603.06961}, year={2026} }}

Figure 1: Sensitivity shaping for reliable out-of-distribution detection in generative dynamics models

CoRL 2026 Sensitivity Shaping for Latent Modeling

Hongzhan Yu, Chenghao Li, Ruipeng Zhang, Henrik Christensen, Sicun Gao

arXiv

BibTeX

@article{yu2026sensitivity, title={Sensitivity Shaping for Latent Modeling}, author={Yu, Hongzhan and Li, Chenghao and Zhang, Ruipeng and Christensen, Henrik and Gao, Sicun}, journal={arXiv preprint arXiv:2606.14585}, year={2026} }

ICML 2025 When Maximum Entropy Misleads Policy Optimization

Ruipeng Zhang, Ya-Chien Chang, Sicun Gao

arXiv Code Slides

BibTeX

@article{zhang2025maximum, title={When maximum entropy misleads policy optimization}, author={Zhang, Ruipeng and Chang, Ya-Chien and Gao, Sicun}, journal={arXiv preprint arXiv:2506.05615}, year={2025} }

ICML 2025 Improving Value Estimation Critically Enhances Vanilla Policy Gradient

Tao Wang, Ruipeng Zhang, Sicun Gao

arXiv Code

BibTeX

@article{wang2025improving, title={Improving value estimation critically enhances vanilla policy gradient}, author={Wang, Tao and Zhang, Ruipeng and Gao, Sicun}, journal={arXiv preprint arXiv:2505.19247}, year={2025} }

CoRL 2025 Sequence Modeling for Time-Optimal Quadrotor Trajectory Optimization with Sampling-based Robustness Analysis

Katherine Mao, Hongzhan Yu, Ruipeng Zhang, Igor Spasojevic, M Ani Hsieh, Sicun Gao, Vijay Kumar

arXiv Code

BibTeX

@article{mao2025sequence, title={Sequence Modeling for Time-Optimal Quadrotor Trajectory Optimization with Sampling-based Robustness Analysis}, author={Mao, Katherine and Yu, Hongzhan and Zhang, Ruipeng and Spasojevic, Igor and Hsieh, M Ani and Gao, Sicun and Kumar, Vijay}, journal={arXiv preprint arXiv:2506.13915}, year={2025} }

NeurIPS 2022 Learning-based motion planning in dynamic environments using GNNs and temporal encoding

Ruipeng Zhang, Chenning Yu, Jingkai Chen, Chuchu Fan, Sicun Gao

arXiv Website Code

BibTeX

@article{zhang2022learning, title={Learning-based motion planning in dynamic environments using gnns and temporal encoding}, author={Zhang, Ruipeng and Yu, Chenning and Chen, Jingkai and Fan, Chuchu and Gao, Sicun}, journal={Advances in Neural Information Processing Systems}, volume={35}, pages={30003--30015}, year={2022} }

Experience

Applied Scientist Intern — Amazon Industrial Robotics Group Jun. 2026 – Sept. 2026

Professional Service

Reviewer: NeurIPS, ICML, ICLR, ICRA, IROS