Can I read Deep CNN for Go Move Evaluation on EtoBox?

Deep CNN for Go Move Evaluation by Sauradeep Debnath is a document available to read on EtoBox.

What is Deep CNN for Go Move Evaluation about?

This paper explores the use of deep convolutional neural networks (CNNs) for move evaluation in the game of Go, demonstrating that a 12-layer CNN can predict expert moves with 55% accuracy, comparable to a 6 dan human player. The CNN outperformed traditional search programs, winning 97% of games against GnuGo and matching the performance of advanced Monte-Carlo tree search programs. The study highlights the potential of deep learning architectures to effectively represent and learn complex evaluation functi

Author

Sauradeep Debnath

Language

EN