Improving Model and Search for Computer Go
Preprint
Abstract
The standard for Deep Reinforcement Learning in games, following Alpha Zero, is to use residual networks and to increase the depth of the network to get better results. We propose to improve mobile networks as an alternative to residual networks and experimentally show the playing strength of the networks according to both their width and their depth. We also propose a generalization of the PUCT search algorithm that improves on PUCT.Keywords
All Related Versions
This publication has 5 references indexed in Scilit:
- Polygames: Improved zero learningICGA Journal: The Journal of the Computer Games Community, 2021
- Mobile Networks for Computer GoIEEE Transactions on Games, 2020
- A general reinforcement learning algorithm that masters chess, shogi, and Go through self-playScience, 2018
- Mastering the game of Go without human knowledgeNature, 2017
- Mastering the game of Go with deep neural networks and tree searchNature, 2016