Analysis and Optimization of Deep CounterfactualValue Networks

07/02/2018
by   Eneldo Loza Mencía, et al.
0

Recently a strong poker-playing algorithm called DeepStack was published, which is able to find an approximate Nash equilibrium during gameplay by using heuristic values of future states predicted by deep neural networks. This paper analyzes new ways of encoding the inputs and outputs of DeepStack's deep counterfactual value networks based on traditional abstraction techniques, as well as an unabstracted encoding, which was able to increase the network's accuracy.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/01/2018

Deep Counterfactual Regret Minimization

Counterfactual Regret Minimization (CFR) is the leading algorithm for so...
research
01/30/2020

Fictitious Play Outperforms Counterfactual Regret Minimization

We compare the performance of two popular iterative algorithms, fictitio...
research
05/26/2021

NNCFR: Minimize Counterfactual Regret with Neural Networks

Counterfactual Regret Minimization (CFR) is the popular method for findi...
research
08/20/2020

Finding the optimal Nash equilibrium in a discrete Rosenthal congestion game using the Quantum Alternating Operator Ansatz

This paper establishes the tractability of finding the optimal Nash equi...
research
05/23/2018

On self-play computation of equilibrium in poker

We compare performance of the genetic algorithm and the counterfactual r...
research
04/01/2021

Back to Square One: Superhuman Performance in Chutes and Ladders Through Deep Neural Networks and Tree Search

We present AlphaChute: a state-of-the-art algorithm that achieves superh...
research
02/28/2022

Neural net modeling of equilibria in NSTX-U

Neural networks (NNs) offer a path towards synthesizing and interpreting...

Please sign up or login with your details

Forgot password? Click here to reset