« earlier | later » Page 1 of 1
[1412.6564] Move Evaluation in Go Using Deep Convolutional Neural Networks edit / delete
"We train a large 12-layer convolutional neural network by supervised learning from a database of human professional games. The network correctly predicts the expert move in 55% of positions, equalling the accuracy of a 6 dan human player."
to ag1084 ai games go neural-networks ... on 05 January 2015
« earlier | later » Page 1 of 1
- neural-networks | |
1 | + ag1084 |
1 | + ai |
1 | + games |
1 | + go |
tasty by Adam Sampson.