Li, Ang

Staff Research Scientist
DeepMind, Mountain View

Research Affiliate
University of Texas, Dallas

Gmail    Twitter   Google Scholar


I am a Research Scientist at Google DeepMind, Mountain View, CA. My research aims at developing deep and reinforced AI systems that are scalable and feasible for learning from the real world. I also work closely with many engineering teams to land cutting-edge AI technologies into Alphabet products. I led the development of Population Based Training into Google's infrastructure. Our infrastructure became the basis for multiple collaborations such as Hyper-parameter Tuning for Google's Self-driving Cars and Automatic Data Augmentation in Waymo. I further landed the application of meta learning in optimizing Graph Neural Networks for Google Maps traffic prediction. I was an AI coach at Google Launchpad Accelerator and consulted a non-profit organization WattTime on applying artificial intelligence with satellite imaging to estimate carbon emission from power plants across the globe.

I received my Ph.D. in Computer Science in 2017 from the University of Maryland College Park, and B.Sc. from Nanjing University. I have also researched artificial intelligence at Carnegie Mellon University, Apple, Google, Facebook AI Research, and Comcast Labs DC.



  • ETA Prediction with Graph Neural Networks in Google Maps
    Austin Derrow-Pinion, Jennifer She, David Wong, Oliver Lange, Todd Hester, Luis Perez, Marc Nunkesser, Seongjae Lee, Xueying Guo, Brett Wiltshire, Peter Battaglia, Vishal Gupta, Ang Li, Zhongwen Xu, Alvaro Sanchez-Gonzalez, Yujia Li, Petar Veličković.
    CIKM 2021
  • Task-agnostic Continual Learning with Hybrid Probabilistic Models
    Polina Kirichenko, Mehrdad Farajtabar, Dushyant Rao, Balaji Lakshminarayanan, Nir Levine, Ang Li, Huiyi Hu, Andrew Gordon Wilson, Razvan Pascanu.
    ICML 2021 Workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models.
  • The Effectiveness of Memory Replay in Large Scale Continual LearningPDF ] [ bibtex ]
    Yogesh Balaji, Mehrdad Farajtabar, Dong Yin, Alex Mott, Ang Li
    CVPR 2021 Workshop on Continual Learning.


  • Learning to Incentivize Other Learning AgentsPDF ] [ bibtex ] [ code ]
    Jiachen Yang, Ang Li, Mehrdad Farajtabar, Peter Sunehag, Edward Hughes and Hongyuan Zha
    NeurIPS 2020.
    - AAMAS 2020 Adaptive and Learning Agents (ALA) Workshop.
  • SOLA: Continual Learning with Second-Order Loss Approximation PDF ] [ bibtex ]
    Dong Yin, Mehrdad Farajtabar, Ang Li
    ArXiv 2020.
    - ICML 2020 Workshop on Continual Learning.
  • The AVA-Kinetics Localized Human Actions Video DatasetPDF ] [ bibtex ]
    Ang Li, Meghana Thotakuri, David A. Ross, Joao Carreira, Alexander Vostrikov, Andrew Zisserman
    ArXiv 2020.
  • Hybrid Models for Open Set RecognitionPDF ] [ Bibtex ]
    Hongjie Zhang, Ang Li, Jie Guo, Yanwen Guo
    ECCV 2020 (Oral).
  • PhysGAN: Generating Physical-World-Resilient Adversarial Examples for Autonomous DrivingPDF ] [ bibtex ]
    Zelun Kong, Junfeng Guo, Ang Li, Cong Liu
    CVPR 2020.
  • Orthogonal Gradient Descent for Continual LearningPDF ] [ Bibtex ]
    Mehrdad Farajtabar, Navid Azizan, Alex Mott, Ang Li
    AISTATS 2020
  • Prediction, Consistency, Curvature: Representation Learning for Locally-Linear ControlPDF ] [ Bibtex ]
    Nir Levine, Yinlam Chow, Rui Shu, Ang Li, Mohammad Ghavamzadeh, Hung Bui
    ICLR 2020
    - Poster at BayLearn 2019.
  • Improved Knowledge Distillation via Teacher AssistantPDF ] [ Bibtex ]
    Seyed-Iman Mirzadeh, Mehrdad Farajtabar, Ang Li, Nir Levine, Akihiro Matsukawa, Hassan Ghasemzadeh
    AAAI 2020.
    - Poster at BayLearn 2019


  • Data Efficient Training for Reinforcement Learning with Adaptive Behavior Policy Sharing
    Ge Liu, Heng-Tze Cheng, Rui Wu, Jing Wang, Jayiden Ooi, Lihong Li, Ang Li, Sibon Li, Craig Boutilier.
    - NeurIPS 2019 Deep Reinforcement Learning Workshop, Vancouver, BC.
    - NeurIPS 2019 Optimization Foundations for Reinforcement Learning Workshop, Vancouver, BC.
  • Cross-View Policy Learning for Street NavigationPDF ] [ Bibtex ]
    Ang Li*, Huiyi Hu*, Piotr Mirowski, Mehrdad Farajtabar
    ICCV 2019.
    - Short paper at ICML 2019 Multitask and Lifelong RL workshop, Long Beach, CA.
    - Poster at BayLearn 2019.
  • Layout-induced Video Representation for Recognizing Agent-in-Place ActionsPDF ] [ Bibtex ]
    Ruichi Yu, Hongcheng Wang, Ang Li, Jingxiao Zheng, Vlad I. Morariu, Larry S. Davis
    ICCV 2019.
    - Poster at BayLearn 2019.
  • A Generalized Framework for Population Based TrainingPDF ] [  Youtube  ] [ Bibtex ]
    Ang Li, Ola Spyra, Sagi Perel, Valentin Dalibard, Max Jaderberg, Chenjie Gu, David Budden, Tim Harley, Pramod Gupta
    KDD 2019.
    - Short paper at NeurIPS 2018 Workshop on Systems for ML, Montréal, Canada.
  • Improving Open Set Domain Adaptation Using Image-to-Image Translation
    Hongjie Zhang, Ang Li, Xu Han, Zhaoming Chen, Yang Zhang, Yanwen Guo
    ICME 2019.
  • SNR: Sub-Network Routing for Flexible Parameter Sharing in Multi-task LearningPDF ]
    Jiaqi Ma, Zhe Zhao, Jilin Chen, Ang Li, Lichan Hong, Ed H. Chi
    AAAI 2019, Honolulu, Hawaii.


  • C-WSL: Count-guided Weakly Supervised LocalizationPDF ] [ Bibtex ]
    Mingfei Gao, Ang Li, Ruichi Yu, Vlad I. Morariu, Larry S. Davis
    ECCV 2018, Munich, Germany.
  • NISP: Pruning Networks using Neuron Importance Score PropagationPDF ] [ Bibtex ]
    Ruichi Yu, Ang Li, Chun-Fu Chen, Jui-Hsin Lai, Vlad I. Morariu, Xintong Han, Mingfei Gao, Ching-Yung Lin, Larry S. Davis
    CVPR 2018, Salt Lake City, Utah.
  • Dynamic Zoom-in Network for Fast Object Detection in Large ImagesPDF ] [ Bibtex ]
    Mingfei Gao, Ruichi Yu, Ang Li, Vlad I. Morariu, Larry S. Davis
    CVPR 2018, Salt Lake City, Utah.







Professional Service

  • Journal Reviewer:
    • IEEE Transaction on Pattern Analysis and Machine Intelligence (TPAMI) 2019, 2020
    • IEEE Transactions on Neural Networks and Learning Systems (TNNLS) 2020
    • IEEE Transaction on Image Processing (TIP) 2017
    • The Visual Computer Journal (TVCJ) 2016
    • SPIE Journal of Electronic Imaging (JEI) 2014, 2016, 2018, 2019
    • IEEE Transaction on Cybernetics 2014
    • Robotics and Autonomous Systems 2018
    • Machine Learning 2019
  • Session Chair:
    • International Joint Conference on Artificial Intelligence (IJCAI) 2021
  • Area Chair:
    • NeurIPS Workshop on Meta Learning 2021
  • Senior Program Committee:
    • AAAI Conference on Artificial Intelligence (AAAI) 2021
    • International Joint Conference on Artificial Intelligence (IJCAI) 2020, 2021
    • International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS) 2020
  • Program Committee or Reviewer:
    • International Conference on Learning Representations (ICLR) 2018, 2019, 2020
    • Annual Conference on Neural Information Processing Systems (NeurIPS) 2017, 2018, 2019
    • IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2017, 2018, 2019
    • International Conference on Machine Learning (ICML) 2017, 2018, 2019
    • IEEE International Conference on Computer Vision (ICCV) 2015, 2017, 2019
    • European Conference on Computer Vision (ECCV) 2018
    • International Joint Conference on Artificial Intelligence (IJCAI) 2019
    • International Conference on Artificial Intelligence and Statistics (AISTATS) 2018, 2020
    • Asian Conference on Computer Vision (ACCV) 2018
    • Asian Conference on Machine Learning (ACML) 2019
    • Winter Conference on Computer Vision (WACV) 2020
    • IEEE International Conference on Biometrics (BTAS) 2012, 2015
    • ACM International Conference on Multimedia (MM) 2014