Scalable Neural Networks for Board Games

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.

• Type: Conference
• Year: 2009
• Where: ICANN
• Authors: Tom Schaul, Jürgen Schmidhuber
• Keywords: ICANN 2009, Multi-Dimensional Recurrent LSTM, Neural Network Architectures, Small Instances