Hanhan Zhou

Did you realize that you are currently living at least one of the dreams that you used to dream?

132.jpeg

hanhan@gwu.edu

Hi! 👋

This is Hanhan, currently a final year Ph.D student at the department of ECE, the George Washington University in the Lab for Intelligent Networking and Computing. I’m fortunate and grateful to be advised by Prof.Tian Lan and worked with Prof.Vaneet Aggarwal and Prof.Guru Venkataramani. I’m interested in Reinforcement Learing, Federated Learning and Generative Models.

My research endeavors have predominantly centered around the design and optimization of decentralized multi-agent reinforcement learning algorithms. This work entails the development of sophisticated methodologies for coordinating agent actions in complex environments without centralized control. Additionally, I have explored the application of reinforcement learning in the domain of network resource allocation to enhance efficiency and performance in distributed systems.

Furthermore, my academic pursuits have included the investigation of heterogeneous federated learning algorithms and their optimizations. In this topic, my focus has been on the optimization of these algorithms to address the challenges posed by the nature of heterogeneous models in distributed environments.

Lastly, my research portfolio also encompasses the domain of offline reinforcement learning, with a particular focus on utilizing sequence generative modeling. This area of study involves the development of generative models that can effectively learn from datasets without interaction with the environment, an approach that holds substantial promise for advancing the field of reinforcement learning by mitigating the dependency on extensive online data collection.

Please contact me if you think I’m a good fit for a position you know or would like to collaborate on research projects, thank you!

Selected Publications

all publications
  1. NeurIPS 2023
    Every Parameter Matters: Ensuring the Convergence of Federated Learning with Dynamic Heterogeneous Models Reduction
    Hanhan Zhou, Tian Lan, Guru Prasadh Venkataramani, and Wenbo Ding
    In Advances in Neural Information Processing Systems, 2023
  2. AAMAS 2023
    MAC-PO: Multi-Agent Experience Replay via Collective Priority Optimization
    Yongsheng Mei, Hanhan Zhou, Tian Lan, Guru Venkataramani, and Peng Wei
    In Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems, 2023
  3. NeurIPS 2022
    PAC: Assisted Value Factorization with Counterfactual Predictions in Multi-Agent Reinforcement Learning
    Hanhan Zhou, Tian Lan, and Vaneet Aggarwal
    In Advances in Neural Information Processing Systems, 2022
  4. IEEE-TETCI
    Value Functions Factorization With Latent State Information Sharing in Decentralized Multi-Agent Policy Gradients
    Hanhan Zhou, Tian Lan, and Vaneet Aggarwal
    IEEE Transactions on Emerging Topics in Computational Intelligence, 2023