Paper Info
Title
Multi-dimensional deep memory Atari-go players for parameter exploring policy gradients
Abstract
Developing superior artificial board-game players is a widely-studied area of Artificial Intelligence. Among the most challenging games is the Asian game of Go, which, despite its deceivingly simple rules, has eluded the development of artificial expert players. In this paper we attempt to tackle this challenge through a combination of two recent developments in Machine Learning. We employ Multi-Dimensional Recurrent Neural Networks with Long Short-Term Memory cells to handle the multi-dimensional data of the board game in a very natural way. In order to improve the convergence rate, as well as the ultimate performance, we train those networks using Policy Gradients with Parameter-based Exploration, a recently developed Reinforcement Learning algorithm which has been found to have numerous advantages over Evolution Strategies. Our empirical results confirm the promise of this approach, and we discuss how it can be scaled up to expert-level Go players.
Year
DOI
Venue
2010
10.1007/978-3-642-15822-3_14
ICANN (2)
Keywords
Field
DocType
artificial expert player,reinforcement learning algorithm,superior artificial board-game player,artificial intelligence,board game,asian game,machine learning,policy gradient,challenging game,multi-dimensional deep memory atari-go,long short-term memory cell,evolution strategies,artificial intelligent,reinforcement learning,neural network,evolution strategy,convergence rate,long short term memory
Multi dimensional,Computer science,Neural network architecture,Recurrent neural network,Rate of convergence,Artificial intelligence,Reinforcement learning algorithm,Machine learning
Conference
Volume
ISSN
ISBN
6353
0302-9743
3-642-15821-8
Citations 
PageRank 
References 
4
0.38
12
Authors
4
Name
Order
Citations
PageRank
Mandy Grüttner140.72
Frank Sehnke252739.18
Tom Schaul391679.40
Jürgen Schmidhuber4178361238.63