Changxi Zhu

Intelligent Systems, Department of Information and Computing Sciences, Utrecht University

Contact: c.zhu@uu.nl

prof_pic.jpg

Buys Ballot Building

Princetonplein 5

Utrecht, Netherlands

I am currently a PhD candidate at the Department of Information and Computing Sciences of Utrecht University working on Multi-Agent Deep Reinforcement Learning with Communication (Comm-MADRL). My supervisors are Prof. Mehdi Dastani and Dr. Shihan Wang. I have a broad interest in Multi-Agent Deep Reinforcement Learning, particularly in how RL agents benefit from communication and interaction with one another.

latest posts

selected publications

  1. Learning Communication Skills in Multi-task Multi-agent Deep Reinforcement Learning
    Changxi Zhu, Mehdi Dastani, and Shihan Wang
    arXiv:2511.03348v2
    Reducing Variance Caused by Communication in Decentralized Multi-agent Deep Reinforcement Learning
    Changxi Zhu, Mehdi Dastani, and Shihan Wang
    arXiv:2502.06261
    Communication with factorized policy gradients in multi-agent deep reinforcement learning
    Changxi Zhu, Mehdi Dastani, and Shihan Wang
    Neural Computing and Applications
    A survey of multi-agent deep reinforcement learning with communication
    Changxi Zhu, Mehdi Dastani, and Shihan Wang
    Autonomous Agents and Multi-Agent Systems