Benjamin Freed
Papers
3
Total Citations
35
H-Index
3
About
Benjamin Freed is a leading researcher in multi-agent reinforcement learning (MARL), with a focus on enabling effective inter-agent communication in complex, partially-observable environments. His major contributions center on developing differentiable communication protocols that allow agents to learn both their policies and their communication strategies simultaneously through backpropagation. This approach, pioneered in his 2020 work "Communication Learning via Backpropagation in Discrete Channels with Unknown Noise" (13 citations), offers significant advantages over traditional methods by converging more quickly to higher-quality cooperative policies. Freed further advanced the field by addressing real-world constraints, such as bandwidth limitations, in his papers "Simultaneous Policy and Discrete Communication Learning for Multi-Agent Cooperation" (11 citations) and "Sparse Discrete Communication Learning for Multi-Agent Cooperation Through Backpropagation" (11 citations). His work bridges the gap between theoretical MARL and practical deployment, making communication both efficient and robust to noisy, discrete channels. Freed's research is essential reading for anyone interested in scalable, communication-aware multi-agent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3