A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning — mentioned by Cybiote

Wed, 06 Dec 2017 13:19:07 GMT

Correction to the above: I stated Deepmind applied Neural Nets+MCTS and achieved ok results. I was actually misremembering two David Silver (Deepmind) papers as one. Smooth UCT modified UCT (popular brand of MCTS) to be able to handle imperfect information games. MCTS does not converge under imperfect information. Smooth UCT is strong at limit poker. Limit is much simpler than no-limit. Neural Fictitious Self Play based on fictitious play (invented 1950s), is an approach to reinforcement learning using neural nets for function approximation. Typical RL methods like DQN are highly exploitable.…

Paper and discussion history. A mention is not an endorsement.

A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning · HackerPapers discussions

A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning — mentioned by Cybiote