Scalable Neural Networks for Board Games
Lists
Only you see this
Abstract
Learning to solve small instances of a problem should help in solving large instances. Unfortunately, most neural network architectures do not exhibit this form of scalability. Our Multi-Dimensional Recurrent LSTM Networks, however, show a high degree of scalability, as we empirically show in the domain of flexible-size board games. This allows them to be trained from scratch up to the level of human beginners, without using domain knowledge.
BibTeX
@inproceedings{schaul2009scalable,
author = {Tom Schaul and Jürgen Schmidhuber},
title = {{Scalable Neural Networks for Board Games}},
booktitle = {ICANN},
year = {2009},
url = {https://openreview.net/forum?id=5UqMmaDNbZ},
}Loading related content...