TY  - CONF
ID  - hopner18DCVNN
T1  - Analysis and Optimization of Deep Counterfactual Value Networks
A1  - Hopner, Patryk
A1  - Loza Mencía, Eneldo
ED  - Trollmann, Frank
ED  - Turhan, Anni-Yasmin
TI  - KI 2018: Advances in Artificial Intelligence
Y1  - 2018
SP  - 305
EP  - 312
PB  - Springer International Publishing
SN  - 978-3-030-00111-7
N1  - Longer version at /bibtex/index.php/publications/show/3118
UR  - /publications/papers/KI18-PokerDCVNN.pdf
M2  - doi: 10.1007/978-3-030-00111-7_26
KW  - deep neural networks
KW  - Game Abstractions
KW  - poker
N2  - 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.
ER  -