Goal recognition is a fundamental cognitive process that enables individuals to infer intentions based on available cues. Current goal recognition algorithms often take only observed actions as input, but in this talk, I'll present a Bayesian framework to explore the role of actions, timing, and goal solvability in goal recognition. This framework provides new insights into human goal recognition and takes a step towards more human-like AI models, fundamental for human-computer interaction. In the second part of the talk, I'll present how planning can be used over the challenging Abstract Reasoning Corpus (ARC), a problem easily solved by humans but highly challenging for both Machine Learning and planning approaches. Both topics will be presented at a high level suited to understanding the core ideas, results and future research opportunities.
Per maggiori informazioni consultare la locandina allegata.