Towards machines that understand people

Andrew Howes; Jussi P. P. Jokinen; A Oulasvirta

doi:10.1002/aaai.12116

Towards machines that understand people

Andrew Howes, Jussi P. P. Jokinen, A Oulasvirta

Research output: Contribution to journal › Article › Scientific › peer-review

10 Citations (Scopus)

59 Downloads (Pure)

Abstract

The ability to estimate the state of a human partner is an insufficient basis on which to build cooperative agents. Also needed is an ability to predict how people adapt their behavior in response to an agent's actions. We propose a new approach based on computational rationality, which models humans based on the idea that predictions can be derived by calculating policies that are approximately optimal given human-like bounds. Computational rationality brings together reinforcement learning and cognitive modeling in pursuit of this goal, facilitating machine understanding of humans.

Original language	English
Pages (from-to)	312-327
Number of pages	16
Journal	AI Magazine
Volume	44
Issue number	3
DOIs	https://doi.org/10.1002/aaai.12116
Publication status	Published - 4 Sept 2023
MoE publication type	A1 Journal article-refereed

Access to Document

10.1002/aaai.12116Licence: CC BY

Towards machines that understand peopleFinal published version, 480 KBLicence: CC BY

Cite this

@article{2fc3edc236614514b953d0971e8f82e7,

title = "Towards machines that understand people",

abstract = "The ability to estimate the state of a human partner is an insufficient basis on which to build cooperative agents. Also needed is an ability to predict how people adapt their behavior in response to an agent's actions. We propose a new approach based on computational rationality, which models humans based on the idea that predictions can be derived by calculating policies that are approximately optimal given human-like bounds. Computational rationality brings together reinforcement learning and cognitive modeling in pursuit of this goal, facilitating machine understanding of humans.",

author = "Andrew Howes and Jokinen, {Jussi P. P.} and A Oulasvirta",

year = "2023",

month = sep,

day = "4",

doi = "10.1002/aaai.12116",

language = "English",

volume = "44",

pages = "312--327",

journal = "AI Magazine",

issn = "0738-4602",

publisher = "John Wiley & Sons",

number = "3",

}

TY - JOUR

T1 - Towards machines that understand people

AU - Howes, Andrew

AU - Jokinen, Jussi P. P.

AU - Oulasvirta, A

PY - 2023/9/4

Y1 - 2023/9/4

N2 - The ability to estimate the state of a human partner is an insufficient basis on which to build cooperative agents. Also needed is an ability to predict how people adapt their behavior in response to an agent's actions. We propose a new approach based on computational rationality, which models humans based on the idea that predictions can be derived by calculating policies that are approximately optimal given human-like bounds. Computational rationality brings together reinforcement learning and cognitive modeling in pursuit of this goal, facilitating machine understanding of humans.

AB - The ability to estimate the state of a human partner is an insufficient basis on which to build cooperative agents. Also needed is an ability to predict how people adapt their behavior in response to an agent's actions. We propose a new approach based on computational rationality, which models humans based on the idea that predictions can be derived by calculating policies that are approximately optimal given human-like bounds. Computational rationality brings together reinforcement learning and cognitive modeling in pursuit of this goal, facilitating machine understanding of humans.

UR - https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=aalto_pure&SrcAuth=WosAPI&KeyUT=WOS:001058084100001&DestLinkType=FullRecord&DestApp=WOS

UR - http://www.scopus.com/inward/record.url?scp=85181478697&partnerID=8YFLogxK

U2 - 10.1002/aaai.12116

DO - 10.1002/aaai.12116

M3 - Article

SN - 0738-4602

VL - 44

SP - 312

EP - 327

JO - AI Magazine

JF - AI Magazine

IS - 3

ER -

Towards machines that understand people

Abstract

Access to Document

Other files and links

Fingerprint

FCAI: Finnish Center for Artificial Intelligence (FCAI)

Human Automata: Simulator-based Methods for Collaborative AI

-: Bayesian Artefact Design

Cite this

Towards machines that understand people

Abstract

Access to Document

Other files and links

Fingerprint

Projects

FCAI: Finnish Center for Artificial Intelligence (FCAI)

Human Automata: Simulator-based Methods for Collaborative AI

-: Bayesian Artefact Design

Cite this