Consistency of feature Markov processes

Peter Sunehag*, Marcus Hutter

*Corresponding author for this work

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    10 Citations (Scopus)

    Abstract

    We are studying long term sequence prediction (forecasting). We approach this by investigating criteria for choosing a compact useful state representation. The state is supposed to summarize useful information from the history. We want a method that is asymptotically consistent in the sense it will provably eventually only choose between alternatives that satisfy an optimality property related to the used criterion. We extend our work to the case where there is side information that one can take advantage of and, furthermore, we briefly discuss the active setting where an agent takes actions to achieve desirable outcomes.

    Original languageEnglish
    Title of host publicationAlgorithmic Learning Theory - 21st International Conference, ALT 2010, Proceedings
    Pages360-374
    Number of pages15
    DOIs
    Publication statusPublished - 2010
    Event21st International Conference on Algorithmic Learning Theory, ALT 2010 - Canberra, ACT, Australia
    Duration: 6 Oct 20108 Oct 2010

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume6331 LNAI
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference21st International Conference on Algorithmic Learning Theory, ALT 2010
    Country/TerritoryAustralia
    CityCanberra, ACT
    Period6/10/108/10/10

    Fingerprint

    Dive into the research topics of 'Consistency of feature Markov processes'. Together they form a unique fingerprint.

    Cite this