[BibTeX] [RIS]
Analysis and Optimization of Deep Counterfactual Value Networks
Type of publication: Inproceedings
Citation: hopner18DCVNN
Booktitle: KI 2018: Advances in Artificial Intelligence
Year: 2018
Pages: 305--312
Publisher: Springer International Publishing
Note: Longer version at https://www.ke.tu-darmstadt.de/bibtex/index.php/publications/show/3118
ISBN: 978-3-030-00111-7
URL: https://www.ke.tu-darmstadt.de/publications/papers/KI18-PokerDCVNN.pdf
DOI: 10.1007/978-3-030-00111-7_26
Abstract: 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.
Keywords: deep neural networks, Game Abstractions, poker
Authors Hopner, Patryk
Loza MencĂ­a, Eneldo
Editors Trollmann, Frank
Turhan, Anni-Yasmin