Modelling Human Decision-making based on Aggregate Observation Data

Antti Kangasrääsiö, Samuel Kaski

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionProfessional

66 Lataukset (Pure)


Being able to infer the goals, preferences and limitations of humans is of key importance in designing interactive systems. Reinforcement learning (RL) models are a promising direction of research, as they are able to model how the behavioural patterns of users emerge from the task and environment structure. One limitation with traditional inference methods for RL models is the strict requirements for observation data; both the states of the environment and the actions of the agent need to be observed at each step of the task. This has prevented RL models from being used in situations where such fine-grained observations are not available. In this extended abstract we present results from a recent study where we demonstrated how inference can be performed for RL models even when the observation data is significantly more coarse-grained.
The idea is to solve the inverse reinforcement learning (IRL) problem using approximate Bayesian computation sped up with Bayesian optimization.
OtsikkoHuman In The Loop-ML Workshop at ICML
TilaJulkaistu - 2017
OKM-julkaisutyyppiD3 Ammatillisen konferenssin julkaisusarja
TapahtumaHuman in the Loop Machine Learning; ICML Workshop - Sydney, Austraalia
Kesto: 11 elokuuta 201711 elokuuta 2017


WorkshopHuman in the Loop Machine Learning; ICML Workshop

Sormenjälki Sukella tutkimusaiheisiin 'Modelling Human Decision-making based on Aggregate Observation Data'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä