Qizhen Wu

Papers

Total Citations

H-Index

About

Qizhen Wu is a leading researcher at the intersection of swarm robotics and multi-agent reinforcement learning, with a focus on enabling intelligent coordination under extreme uncertainty. His work addresses two fundamental challenges in the field: training efficiency and decision-making in adversarial environments. In his highly cited 2024 paper, “Hierarchical Reinforcement Learning for Swarm Confrontation With High Uncertainty,” Wu tackles the complex hybrid decision processes inherent in pursuit-evasion games, where unknown opponent strategies and dynamic obstacles create a high-stakes, partially observable environment. This work has already garnered 9 citations, signaling its immediate impact. Building on this, his 2025 paper, “Lyapunov-Informed Multi-Agent Reinforcement Learning for Multi-Robot Cooperation Tasks,” introduces a novel approach to the pervasive problem of low training efficiency in MARL. By integrating Lyapunov stability theory—a concept from control theory—as prior knowledge, Wu demonstrates how environmental properties can be leveraged to dramatically accelerate learning and improve convergence. This innovative synthesis of control-theoretic guarantees with deep reinforcement learning marks a significant contribution, positioning Wu as a rising figure in the quest for robust, scalable, and sample-efficient multi-robot systems.