Bolei Zhou

Assistant Professor
Computer Science Department, University of California, Los Angeles
Office: 295D, Engineering VI, UCLA

I study interpretable human-AI interaction for computer vision and machine autonomy. I am also interested in understanding various human-centric properties of current AI models beyond their accuracy, such as explainability, interpretability, steerability, generalization, and safety. Some of the earlier works I co-authored are Class Activation Mapping (CAM), Places, ADE20K, Network Dissection.

See MetaDriverse for recent work on machine autonomy and GenForce for recent work on generative modeling.

News

Sep 12, 2023 Thank Intel for supporting our research via Intel’s 2023 Rising Star Faculty Award.
Apr 26, 2023 Invited talks at workshops on coPerception: collaborative perception and learning at ICRA’23, end-to-end autonomous driving at CVPR’23, and secure and safe autonomous driving at CVPR’23, and BIRS workshop on 3D generative models.
Mar 24, 2023 Grateful to receive NSF award for supporting our research of developing our MetaDrive driving simulator into MetaDriverse, an open-source infrastructure for AI research on autonomous driving.
Jan 14, 2023 This NeurIPS’22 workshop talk Toward Safe and Generalizable Embodied AI for Driving Decision-Making highlights some of our recent effort on machine autonomy and embodied AI.
Jan 9, 2023 I summarize our effort of going from Network Dissection to Policy Dissection in the talk Discovering Interpretable Concepts in Deep Representations at IPAM Workshop on Explainable AI for the Sciences: Towards Novel Insights.
Dec 5, 2022 I finally start collecting my talks recorded over the years in order to create digital twin of myself:).
Nov 12, 2022 Invited talks at the sunny SoCal NLP Symposium 2022 in UCSB and NeurIPS’22 Workshop on ML for Autonomous Driving in New Orleans, in person.
Oct 15, 2022 Thank Amazon for supporting our research via Amazon Research Award.
Jun 20, 2022 1st Tutorial on Human-Centered AI for Computer Vision is hosted at CVPR’22 in New Orleans. Slides are available there. I gave a talk on Human-AI interaction for computer vision.

Selected Publications

  1. NeurIPS Spotlight
    Learning from Active Human Involvement through Proxy Value Propagation
    Zhenghao Peng, Wenjie Mo, Chenda Duan, Quanyi Li, and Bolei Zhou
    In Neural Information Processing Systems (NeurIPS Spotlight) , 2023
  2. NeurIPS
    ScenarioNet: Open-Source Platform for Large-Scale Traffic Scenario Simulation and Modeling
    Quanyi Li, Zhenghao Peng, Lan Feng, Zhizheng Liu, Chenda Duan, Wenjie Mo, and Bolei Zhou
    Neural Information Processing Systems, Datasets and Benchmarks (NeurIPS) , 2023
  3. CoRL
    CAT: Closed-loop Adversarial Training for Safe End-to-End Driving
    Linrui Zhang, Zhenghao Peng, Quanyi Li, and Bolei Zhou
    In 7th Annual Conference on Robot Learning (CoRL) , 2023
  4. ICCV
    One-shot generative domain adaptation
    Ceyuan Yang, Yujun Shen, Zhiyi Zhang, Yinghao Xu, Jiapeng Zhu, Zhirong Wu, and Bolei Zhou
    International Conference on Computer Vision (ICCV) , 2023
  5. CVPR Highlight
    DisCoScene: Spatially Disentangled Generative Radiance Fields for Controllable 3D-aware Scene Synthesis
    Yinghao Xu, Menglei Chai, Zifan Shi, Sida Peng, Ivan Skorokhodov, Aliaksandr Siarohin, Ceyuan Yang, Yujun Shen, Hsin-Ying Lee, Bolei Zhou, and Sergey Tulyakov
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR Highlight) , 2023
  6. ICLR Spotlight
    Guarded Policy Optimization with Imperfect Online Demonstrations
    Zhenghai Xue, Zhenghao Peng, Quanyi Li, Zhihan Liu, and Bolei Zhou
    International Conference on Learning Representations (ICLR Spotlight) , 2023
  7. ICRA
    TrafficGen: Learning to Generate Diverse and Realistic Traffic Scenarios
    Lan Feng, Quanyi Li, Zhenghao Peng, Shuhan Tan, and Bolei Zhou
    IEEE International Conference on Robotics and Automation (ICRA) , 2023
  8. NeurIPS
    Human-AI Shared Control via Policy Dissection
    Quanyi Li, Zhenghao Peng, Haibin Wu, Lan Feng, and Bolei Zhou
    Neural Information Processing Systems (NeurIPS) , 2022
  9. ECCV
    Learning to Drive by Watching YouTube Videos: Action-Conditioned Contrastive Policy Pretraining
    Qihang Zhang, Zhenghao Peng, and Bolei Zhou
    European Conference on Computer Vision (ECCV) , 2022
  10. TPAMI
    MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning
    Quanyi Li*, Zhenghao Peng*, Lan Feng, Qihang Zhang, Zhenghai Xue, and Bolei Zhou
    IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) , 2022
  11. CVPR
    3D-aware Image Synthesis via Learning Structural and Textural Representations
    Yinghao Xu, Sida Peng, Ceyuan Yang, Yujun Shen, and Bolei Zhou
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , 2022
  12. ICLR
    Efficient Learning of Safe Driving Policy via Human-AI Copilot Optimization
    Quanyi Li, Zhenghao Peng, and Bolei Zhou
    International Conference on Learning Representations (ICLR) , 2022