We present a system that can learn to represent actions aswell as to internally simulate the likely continuation of their initial parts.The method we propose is based on the Associative Self Organizing Map(A-SOM), a variant of the Self Organizing Map. By emulating the waythe human brain is thought to perform pattern recognition tasks, the A-SOM learns to associate its activity with di erent inputs over time, whereinputs are observations of other's actions. Once the A-SOM has learnt torecognize actions, it uses this learning to predict the continuation of anobserved initial movement of an agent, in this way reading its intentions.We evaluate the system's ability to simulate actions in an experimentwith good results, and we provide a discussion about its generalizationability. The presented research is part of a bigger project aiming at en-dowing an agent with the ability to internally represent action patternsand to use these to recognize and simulate others behaviour.
|Numero di pagine||13|
|Stato di pubblicazione||Published - 2013|
All Science Journal Classification (ASJC) codes