Deep Reinforcement Learning from Self-Play in Imperfect-Information Games

By Johannes Heinrich, David Silver. 2016

468 citations recorded by the source corpus.

Many real-world applications can be described as large-scale games of imperfect information. To deal with these challenging domains, prior work has focused on computing Nash equilibria in a handcrafted abstraction of the domain. In this paper we introduce the first scalable end-to-end approach to learning approximate Nash equilibria without prior domain knowledge. Our method combines fictitious self-play with deep reinforcement learning. When applied to Leduc poker, Neural Fictitious Self-Play (NFSP) approached a Nash equilibrium, whereas common reinforcement learning methods diverged. In Limit Texas Holdem, a poker game of real-world scale, NFSP learnt a strategy that approached the performance of state-of-the-art, superhuman algorithms based on significant domain expertise.

Subfields: Artificial Intelligence, Reinforcement Learning, Game & Decision Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Sergey Levine
  2. Pieter Abbeel
  3. Yoshua Bengio
  4. Ruslan Salakhutdinov
  5. Yuxi Li
  6. Mohit Bansal
  7. Shimon Whiteson
  8. Quoc V. Le

In Reinforcement Learning

  1. Sergey Levine
  2. Pieter Abbeel
  3. Shimon Whiteson
  4. Yuxi Li
  5. Jakob N. Foerster
  6. Chelsea Finn
  7. David Silver
  8. Timothy P. Lillicrap

In Game & Decision Theory

  1. Noam Brown
  2. Tuomas Sandholm
  3. Neil Burch
  4. Marc Lanctot
  5. Karl Tuyls
  6. Julien Pérolat
  7. Yuxi Li
  8. Adam Lerer

Citations in the corpus