ANU Open Research Repository has been upgraded. We are still working on a few minor issues, which may result in short outages throughout the day. Please get in touch with repository.admin@anu.edu.au if you experience any issues.
 

Feature reinforcement learning: state of the art

Date

2014-07-28

Authors

Daswani, Mayank
Sunehag, Peter
Hutter, Marcus

Journal Title

Journal ISSN

Volume Title

Publisher

Association for the Advancement of Artificial Intelligence

Abstract

Feature reinforcement learning was introduced five years ago as a principled and practical approach to history-based learning. This paper examines the progress since its inception. We now have both model-based and model-free cost functions, most recently extended to the function approximation setting. Our current work is geared towards playing ATARI games using imitation learning, where we use Feature RL as a feature selection method for high-dimensional domains.

Description

Keywords

Reinforcement learning, temporal difference learning, partial observability, Q-learning, feature learning, function approximation, rational agents

Citation

Source

Type

Conference paper

Book Title

Sequential decision-making with big data: papers from the AAAI-14 workshop

Entity type

Access Statement

License Rights

DOI

Restricted until