no code implementations • 19 Apr 2024 • Douglas Rebstock, Christopher Solinas, Nathan R. Sturtevant, Michael Buro
Traditional search algorithms have issues when applied to games of imperfect information where the number of possible underlying states and trajectories are very large.
1 code implementation • 3 Mar 2021 • Varun Bhatt, Michael Buro
In this paper, we first study learning in matrix-based signaling games to empirically show that decentralized methods can converge to a suboptimal policy.
no code implementations • 14 Dec 2020 • Jake Tuero, Michael Buro
It can also quantify its uncertainty in its predictions, allowing for algorithm portfolio models to make better informed decisions about which algorithm to run on a particular instance.
no code implementations • 22 May 2020 • Arta Seify, Michael Buro
The combination of Monte-Carlo Tree Search (MCTS) and deep reinforcement learning is state-of-the-art in two-player perfect-information games.
no code implementations • 27 May 2019 • Douglas Rebstock, Christopher Solinas, Michael Buro, Nathan R. Sturtevant
Trick-taking card games feature a large amount of private information that slowly gets revealed through a long sequence of actions.
no code implementations • 27 May 2019 • Douglas Rebstock, Christopher Solinas, Michael Buro
In trick-taking card games like Bridge or Skat, large information sets and an inability to advance the simulation without fully determinizing the state make forward search problematic.
no code implementations • 22 Mar 2019 • Christopher Solinas, Douglas Rebstock, Michael Buro
In trick-taking card games, a two-step process of state sampling and evaluation is widely used to approximate move values.
no code implementations • 11 Sep 2017 • Nicolas A. Barriga, Marius Stanescu, Michael Buro
The network is then used to select a script --- an abstract action --- to produce low level actions for all units.