We propose a system able to represent others’ actions as well as to internally simulate their likely continuationfrom a partial observation. The approach presented here is the first step towards a more ambitious goalof endowing an artificial agent with the ability to recognise and predict others’ intentions. Our approachis based on the associative self-organising map, a variant of the self-organising map capable of learningto associate its activity with different inputs over time, where inputs are processed observations of others’actions. We have evaluated our system in two different experimental scenarios obtaining promisingresults: the system demonstrated an ability to learn discriminable representations of actions, to recognisenovel input, and to simulate the likely continuation of partially seen actions.
|Numero di pagine||20|
|Stato di pubblicazione||Published - 2015|
All Science Journal Classification (ASJC) codes