Macheng Shen
Papers
1
Total Citations
17
H-Index
1
About
Macheng Shen is a leading researcher in multiagent reinforcement learning (MARL) and robotic systems, with a focus on tackling the critical scalability challenges that arise when coordinating large teams of autonomous agents. His most cited work, "Scaling Up Multiagent Reinforcement Learning for Robotic Systems: Learn an Adaptive Sparse Communication Graph" (2020, 17 citations), addresses the exponential complexity that plagues MARL in multiagent systems. Shen’s key contribution lies in developing an adaptive, sparse communication graph that allows agents to selectively share information, dramatically reducing computational overhead while preserving coordination quality. This innovation enables MARL to be applied to large-scale robotic swarms and distributed systems where traditional methods fail. By demonstrating that agents can learn when and with whom to communicate, Shen has opened new pathways for deploying reinforcement learning in real-world multirobot tasks, from warehouse automation to search-and-rescue missions. His work is widely recognized for bridging the gap between theoretical MARL and practical, scalable deployment, making him a pivotal figure in advancing autonomous multiagent systems.
Research Focus
Key Achievements
Top Papers
- 1