“Despite significant breakthroughs in artificial intelligence, it has been notoriously hard for computers to understand why a user behaves the way she does. Cognitive models that describe individual capabilities, as well as goals, can much better explain and hence be able to predict individual behaviour also in new circumstances. However, learning these models from the practically available indirect data has been out of reach,” said the University.
That was before the researchers applied approximate Bayesian computation (ABC), a machine learning method, used in climate sciences and epidemiology, that infers complex models from observations.
According to Aalto, just by observing how long a user takes to click menu items, one can infer a model that reproduces similar behaviour and accurately estimates some characteristics of that user’s visual system, such as fixation durations.
“The benefit of our approach is that much smaller amount of data is needed than for ‘black box’ methods,” said Aalto researcher Antti Kangasrääsiö. “Previous methods for performing this type of tuning have either required extensive manual labour, or a large amount of very accurate observation data, which has limited the applicability of these models.”
Apparently, this could be useful in human-robot interaction, or in assessing individual capabilities automatically, for example detecting symptoms of cognitive decline.
“We will be able to infer a model of a person that also simulates how that person learns to act in totally new circumstances,” said fellw researcher Professor Samuel Kaski
Aalto University worked with the Universities of Birmingham and Oslo. Results will be presented at computer-human interaction conference CHI in Denver.