Modelling Human Decision-making based on Aggregate Observation Data

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

62 Downloads (Pure)

Abstract

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.
Original languageEnglish
Title of host publicationHuman In The Loop-ML Workshop at ICML
Place of PublicationSydney
Number of pages4
Publication statusPublished - 2017
MoE publication typeD3 Professional conference proceedings
EventHuman in the Loop Machine Learning; ICML Workshop - Sydney, Australia
Duration: 11 Aug 201711 Aug 2017
https://machlearn.gitlab.io/hitl2017/

Workshop

WorkshopHuman in the Loop Machine Learning; ICML Workshop
Abbreviated titleHITL
CountryAustralia
CitySydney
Period11/08/201711/08/2017
Internet address

Fingerprint Dive into the research topics of 'Modelling Human Decision-making based on Aggregate Observation Data'. Together they form a unique fingerprint.

Cite this