Mixed-Policy Asynchronous Deep Q-Learning

• Type: Article in International Conference Proceedings Book
• Date: 2017

Authors

Simões, D (Author) Other; lau, n (Author) FCUP; reis, lp (Author) REIT

Conference proceedings International

Title: Advances in Intelligent Systems and Computing

Pages: 129-140

3rd Iberian Robotics Conference, ROBOT 2017

22 November 2017 through 24 November 2017

Indexing

Scopus - 8 Citations

Other information

Authenticus ID: P-00N-N4Z

DOI: 10.1007/978-3-319-70836-2_11

Abstract (EN): There are many open issues and challenges in the reinforcement learning field, such as handling high-dimensional environments. Function approximators, such as deep neural networks, have been successfully used in both single- and multi-agent environments with high dimensional state-spaces. The multi-agent learning paradigm faces even more problems, due to the effect of several agents learning simultaneously in the environment. One of its main concerns is how to learn mixed policies that prevent opponents from exploring them in competitive environments, achieving a Nash equilibrium. We propose an extension of several algorithms able to achieve Nash equilibriums in single-state games to the deep-learning paradigm. We compare their deep-learning and table-based implementations, and demonstrate how WPL is able to achieve an equilibrium strategy in a complex environment, where agents must find each other in an infinite-state game and play a modified version of the Rock Paper Scissors game. © Springer International Publishing AG 2018.

Language: English

Type (Professor's evaluation): Scientific