Skip navigation
Skip navigation

Feature Reinforcement Learning: Part I. unstructured MDPs

Hutter, Marcus

Description

General-purpose, intelligent, learning agents cycle through sequences of observations, actions, and rewards that are complex, uncertain, unknown, and non-Markovian. On the other hand, reinforcement learning is well-developed for small finite state Markov decision processes (MDPs). Up to now, extracting the right state representations out of bare observations, that is, reducing the general agent setup to the MDP framework, is an art that involves significant effort by designers. The...[Show more]

CollectionsANU Research Publications
Date published: 2009-10
Type: Journal article
URI: http://hdl.handle.net/1885/14919
Source: Journal of Artificial General Intelligence
DOI: 10.2478/v10229-011-0002-8

Download

File Description SizeFormat Image
Hutter Feature Reinforcement Learning 2009.pdf404.5 kBAdobe PDFThumbnail


Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.

Updated:  12 November 2018/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator