Changxi Zhu

I received my Ph.D. in Computer Science from Utrecht University on 29 October 2026. My dissertation, Learning with Communication in Multi-agent Deep Reinforcement Learning, was completed at Intelligent Systems, Department of Information and Computing Sciences, Utrecht University under the supervision of Prof. Mehdi Dastani and Dr. Shihan Wang.

My research develops adaptive and learnable communication methods for multi-agent deep reinforcement learning in stochastic, continuous-control, and multi-task environments. I am interested in making multi-agent systems more efficient, robust, and generalizable through communication and interaction.