Counterfactual Multi-Agent Policy Gradients
About this video
Advanced · Hand-picked by a BrighterTube curator · found via round-3 back catalogue of a channel Lumen picks video by video · 82% confidenceCounterfactual multi-agent policy gradients (49 min) teaches reinforcement learning technique for multi-agent systems. Covers algorithmic approach to credit assignment in decentralized control.
A 49-minute video from Microsoft Research: “Counterfactual Multi-Agent Policy Gradients”.
Video by Microsoft Research, played through YouTube's embedded player.