Inference of Strategic Behavior based on Incomplete Observation Data

Antti Kangasrääsiö, Samuel Kaski

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionProfessional

100 Lataukset (Pure)

Abstrakti

Inferring the goals, preferences and restrictions of strategically behaving agents is a common goal in many situations, and an important requirement for enabling computer systems to better model and understand human users.
Inverse reinforcement learning (IRL) is one method for performing this kind of inference based on observations of the agent's behavior.
However, traditional IRL methods are only applicable when the observations are in the form of state-action paths -- an assumption which does not hold in many real-world modelling settings.
This paper demonstrates that inference is possible even with an arbitrary observation noise model.
AlkuperäiskieliEnglanti
OtsikkoNIPS17 Workshop: Learning in the Presence of Strategic Behavior
KustantajaCarnegie Mellon University
Sivumäärä4
TilaJulkaistu - 8 jouluk. 2017
OKM-julkaisutyyppiD3 Ammatillisen konferenssin julkaisusarja
TapahtumaANNUAL CONFERENCE ON NEURAL INFORMATION PROCESSING SYSTEMS - Long Beach, Yhdysvallat
Kesto: 4 jouluk. 20179 jouluk. 2017
Konferenssinumero: 31

Conference

ConferenceANNUAL CONFERENCE ON NEURAL INFORMATION PROCESSING SYSTEMS
LyhennettäNIPS
Maa/AlueYhdysvallat
KaupunkiLong Beach
Ajanjakso04/12/201709/12/2017

Sormenjälki

Sukella tutkimusaiheisiin 'Inference of Strategic Behavior based on Incomplete Observation Data'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä