More

Pronouns: he/him/his

Education

University of California, Los Angeles

• September 2022 - Present
• PhD student at Department of Computer Science.
• Supervised by Professor Bolei Zhou.

The Chinese University of Hong Kong

• August 2019 - July 2022
• MPhil student at Department of Information Engineering.
• Supervised by Professor Bolei Zhou.

Shanghai Jiao Tong University

• September 2015 - July 2019
• Bachelor of Engineering and member of Zhiyuan Honors Program.

Teaching

Awards

• Amazon Fellowship, 2024-25
• University Fellowship, 2023-24
• Outstanding Tutors Award 2021 of the Faculty of Engineering, CUHK
• Teaching Assistant Awards, Term 2, 2020-21
• Teaching Assistant Awards, Term 1, 2020-21

Teaching Assistance

• CS260R Reinforcement Learning at UCLA, Fall 23
• CS269 Seminar on Reinforcement Learning at UCLA, Fall 22
• IERG5350 Reinforcement Learning at CUHK, Term 1, 2021-22
• CSCI2100E Data Structures at CUHK, Term 2, 2020-21
• IERG5350 Reinforcement Learning at CUHK, Term 1, 2020-21
• IERG6130 Seminar on Reinforcement Learning at CUHK, Term 2, 2019-20

Press Coverage

• A group of journalist visited our Embodied AI lab at CPII On December 6, 2021. The following press make their reports: BastillePost 巴士的報, HK01 香港01, HK Commercial Daily 香港商報, South China Morning Post, Sing Tao Daily 星島日報, HK Economic Journal 信報, Ta Kung Pao 大公报.

Papers

• Zhenghao Peng, Wenjie Luo, Yiren Lu, Tianyi Shen, Cole Gulino, Ari Seff, and Justin Fu. Improving agent behaviors with rl fine-tuning for autonomous driving. European Conference on Computer Vision, 2024 (ECCV 2024)[PDF]
• Yunsong Zhou, Michael Simon, Zhenghao Peng, Sicheng Mo, Hongzi Zhu, Minyi Guo, and Bolei Zhou. SimGen: Simulator-conditioned driving scene generation. Advances in Neural Information Processing Systems, 2024 (NeurIPS 2024)[PDF, Website]
• Brandon J. McMahan, Zhenghao Peng, Bolei Zhou, and Jonathan C. Kao. Shared autonomy with: Interventional diffusion assistance. Advances in Neural Information Processing Systems, 2024 (NeurIPS 2024) [ PDF ]

• Zhenghao Peng, Wenjie Mo, Chenda Duan, Quanyi Li, and Bolei Zhou. Learning from active human involvement through proxy value propagation. Advances in Neural Information Processing Systems, 2023 (NeurIPS 2023 Spotlight) [ PDF, Website ]

• Quanyi Li*, Zhenghao Peng*, Lan Feng, Zhizheng Liu, Chenda Duan, Wenjie Mo, and Bolei Zhou. Scenarionet: Open-source platform for large-scale traffic scenario simulation and modeling. Advances in Neural Information Processing Systems, 2023 (NeurIPS 2023) [ PDF , Code , Website ]

• Linrui Zhang, Zhenghao Peng, Quanyi Li, and Bolei Zhou. Cat: Closed-loop adversarial training for safe end-to- end driving. In 7th Annual Conference on Robot Learning, 2023 (CoRL 2023) [ PDF, Code, Website ]

• Lan Feng, Quanyi Li, Zhenghao Peng*, Shuhan Tan, Bolei Zhou. TrafficGen: Learning to Generate Diverse and Realistic Traffic Scenarios (ICRA 2023) [Webpage] [PDF]

• Zhenghai Xue, Zhenghao Peng, Quanyi Li, Zhihan Liu, Bolei Zhou. Guarded Policy Optimization with Imperfect Online Demonstrations. (ICLR 2023) [OpenReview]

• Qihang Zhang, Zhenghao Peng, Bolei Zhou. Action-Conditioned Contrastive Policy Pretraining. (ECCV 2022) [Webpage] [PDF]

• Quanyi Li, Zhenghao Peng, Haibin Wu, Lan Feng, Bolei Zhou. Human-AI Shared Control via Frequency-based Policy Dissection. (NeurIPS 2022) [Webpage] [PDF]

• Hao Sun, Zhenghao Peng, Bo Dai, Jian Guo, Dahua Lin, and Bolei Zhou. Novel Policy Seeking with Constrained Optimization. (Deep RL Workshop NeurIPS 2022) [PDF]

• Hao Sun, Ziping Xu, Zhenghao Peng, Meng Fang, Bo Dai, Bolei Zhou. MOPA: a Minimalist Off-Policy Approach to Safe-RL. (Deep RL Workshop NeurIPS 2022)

• Quanyi Li*, Zhenghao Peng*, Lan Feng, Qihang Zhang, Zhenghai Xue, Bolei Zhou. MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning. (TPAMI) [Webpage] [PDF]

• Boli Fang, Zhenghao Peng, Hao Sun, and Qin Zhang. Meta Proximal Policy Optimization for Cooperative Multi-gent Continuous Control. (IJCNN 2022)

• Mingxin Huang, Yuliang Liu, Zhenghao Peng, Chongyu Liu, Dahua Lin, Shenggao Zhu, Nicholas Yuan, Kai Ding, and Lianwen Jin. Swintextspotter: Scene text spotting via better synergy between text detection and text recognition. (CVPR 2022)

• Quanyi Li*, Zhenghao Peng*, and Bolei Zhou. Efficient Learning of Safe Driving Policy via Human-AI Copilot Optimization. (ICLR 2022) [Webpage] [PDF]

• Zhenghao Peng*, Quanyi Li*, Chunxiao Liu, and Bolei Zhou. Safe Driving via Expert Guided Policy Optimization. (CoRL 2021) [Webpage] [PDF]

• Zhenghao Peng, Quanyi Li, Chunxiao Liu, and Bolei Zhou. Learning to Simulate Self-driven Particles System with Coordinated Policy Optimization. (NeurIPS 2021) [Webpage] [PDF]

• Quanyi Li*, Zhenghao Peng*, Qihang Zhang, Chunxiao Liu, and Bolei Zhou. Improving the Generalization of End-to-end Driving through Procedural Generation. (arXiv preprint) [PDF]

• Zhenghao Peng, Hao Sun, and Bolei Zhou. Non-local Policy Optimization via Diversity-regularized Collaborative Exploration. (arXiv preprint) [PDF]

• Zhuoran Song, Dongyu Ru, Ru Wang, Hongru Huang, Zhenghao Peng, Jing Ke, Xiaoyao Liang, and Li Jiang. Approximate Random Dropout. Design, Automation & Test in Europe Conference & Exhibition 2019

• Zhenghao Peng, Xuyang Chen, Chengwen Xu, Naifeng Jing, Xiaoyao Liang, Cewu Lu, and Li Jiang. AXNet: Approximate Computing Using an End-to-end Trainable Neural Network. International Conference on Computer-Aided Design 2018