Deep Reinforcement Learning from Self-Play in Imperfect-Information Games · HackerPapers discussions

Deep Reinforcement Learning from Self-Play in Imperfect-Information Games — mentioned by thomasahle

Mon, 13 Dec 2021 22:08:31 GMT

I'm more surprised David Silver isn't on it, since his background is in imperfect information games, with papers such as https://arxiv.org/abs/1603.01121 He did multiple poker papers before he was the main author of Alpha Zero.

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

Deep Reinforcement Learning from Self-Play in Imperfect-Information Games — mentioned by sanxiyn

Thu, 14 Dec 2017 08:41:54 GMT

I would consider a successful application of AlphaZero algorithm to imperfect information games to be a separate achievement, not a done deal. David Silver himself, who led AlphaGo project, wrote in 2016 in "Deep Reinforcement Learning from Self-Play in Imperfect-Information Games" https://arxiv.org/abs/1603.01121 "While many machine learning methods have achieved near-optimal solutions to classical, perfect-information games, these methods fail to converge in imperfect-information games".

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

Deep Reinforcement Learning from Self-Play in Imperfect-Information Games — mentioned by nl

Mon, 16 Jan 2017 00:10:24 GMT

That paper doesn't mention game theory or the Nash equilibrium at all. Perhaps you mean [2], which does. However, the article seems to misinterpret this: the fact that that system approximates the Nash equilibrium should be expected. Any system (including humans) that performs well in those types of games will, almost by definition. [2] https://arxiv.org/pdf/1603.01121.pdf

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

Deep Reinforcement Learning from Self-Play in Imperfect-Information Games — HN submission

Fri, 01 Apr 2016 11:51:53 GMT

The lead researcher behind AlphaGo describing how to build a poker bot

7 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.