University of Hertfordshire

By the same authors

On information-optimal scripting of actions

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Documents

View graph of relations
Original languageEnglish
Title of host publicationAlife 2019
Subtitle of host publicationProceedings of the Artificial Life Conference 2019
EditorsHarold Fellermann, Jaume Bacardit, Ángel Goñi-Moreno, Rudolf Füchslin
PublisherMIT Press
Pages624-625
Number of pages2
DOIs
Publication statusPublished - Jul 2019
EventThe 2019 Conference on Artificial Life - Newcastle, United Kingdom
Duration: 29 Jul 20192 Aug 2019
https://2019.alife.org/

Conference

ConferenceThe 2019 Conference on Artificial Life
Country/TerritoryUnited Kingdom
CityNewcastle
Period29/07/192/08/19
Internet address

Abstract

Animals and humans encounter many tasks which permit ritualized behaviours, essentially fixed action sequences or “scripts”, similar to options known from Reinforcement Learning, but proceeding without intermediate decisions. While running a script, they proceed in an open-loop fashion. However even when these are already known, an agent needs to decide whether to perform a basic action or to trigger a script regarding the particular task. Here we study if including such scripts (i.e. behaviour rituals) is advantageous from the point of view of the relevant information required to take the decision to start such a script depending on the tasks. To achieve this, we modify the relevant information framework including sequences of basic actions to the possible actions.

Notes

Best paper award.

ID: 16789194