Please use this identifier to cite or link to this item:
Title: "Superstition" in the network : deep reinforcement learning plays deceptive games
Authors: Bontrager, Philip
Khalifa, Ahmed
Anderson, Damien
Stephenson, Matthew
Salge, Christoph
Togelius, Julian
Keywords: Computer games
Artificial intelligence
Machine learning
Reinforcement learning
Issue Date: 2019
Publisher: Association for the Advancement of Artificial Intelligence (AAAI)
Citation: Bontrager, P., Khalifa, A., Anderson, D., Stephenson, M., Salge, C., & Togelius, J. (2019). "Superstition" in the network : deep reinforcement learning plays deceptive games. Proceedings of the Fifteenth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-19), Atlanta. 10-16.
Abstract: Deep reinforcement learning has learned to play many games well, but failed on others. To better characterize the modes and reasons of failure of deep reinforcement learners, we test the widely used Asynchronous Actor-Critic (A2C) algorithm on four deceptive games, which are specially designed to provide challenges to game-playing agents. These games are implemented in the General Video Game AI framework, which allows us to compare the behavior of reinforcement learningbased agents with planning agents based on tree search. We find that several of these games reliably deceive deep reinforcement learners, and that the resulting behavior highlights the shortcomings of the learning algorithm. The particular ways in which agents fail differ from how planning-based agents fail, further illuminating the character of these algorithms. We propose an initial typology of deceptions which could help us better understand pitfalls and failure modes of (deep) reinforcement learning.
ISSN: 23340924
Appears in Collections:Scholarly Works - InsDG

Files in This Item:
File Description SizeFormat 
  Restricted Access
1.11 MBAdobe PDFView/Open Request a copy

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.