Reactive Layer in AGI Agent

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F11%3A43898519" target="_blank" >RIV/49777513:23520/11:43898519 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Reactive Layer in AGI Agent

Popis výsledku v původním jazyce

This article presents basic mechanisms of cognition working in AGI agent. I argue that reactive behavior is the baseline of intelligence ? it is the base component working and it can be further extended to produce more intelligent agents. Mechanisms employed at reactive level enable the agent to develop behavior which both explores and exploits the environment with the purpose of receiving highest reward possible. Three fundamental mechanisms are intertwined ? action selection, action value estimation and situation discrimination. Whole process of adaptation is completely unsupervised and depends only on reward received from environment. Some technical details of implementation of given mechanisms (BAGIB agent) are described together with implicationsto other planned parts of ?AGI-compliant? architecture. Discussed are several challenges we encounter in AGI, which are not present in usually narrow and domain-limited approach to AI.

Název v anglickém jazyce

Reactive Layer in AGI Agent

Popis výsledku anglicky

This article presents basic mechanisms of cognition working in AGI agent. I argue that reactive behavior is the baseline of intelligence ? it is the base component working and it can be further extended to produce more intelligent agents. Mechanisms employed at reactive level enable the agent to develop behavior which both explores and exploits the environment with the purpose of receiving highest reward possible. Three fundamental mechanisms are intertwined ? action selection, action value estimation and situation discrimination. Whole process of adaptation is completely unsupervised and depends only on reward received from environment. Some technical details of implementation of given mechanisms (BAGIB agent) are described together with implicationsto other planned parts of ?AGI-compliant? architecture. Discussed are several challenges we encounter in AGI, which are not present in usually narrow and domain-limited approach to AI.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

<a href="/cs/project/ME%20949" target="_blank" >ME 949: Analysis of negative impacts on driver attention</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2011

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

ICAART 2011

ISBN

978-989-8425-40-9

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

721-726

Název nakladatele

SciTePress - Science and Technology Publications

Místo vydání

Setúbal

Místo konání akce

Řím, Itálie

Datum konání akce

28. 1. 2011

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—