22nd Workshop "From Objects to Agents"
Toggle Dark/Light/Auto mode Toggle Dark/Light/Auto mode Toggle Dark/Light/Auto mode

Fabio Bellifemine Keynote

Bounded and Resource-Rational Agents for Integrated Intelligence

keynote speech by Antonio Lieto

Abstract

In this talk, I will present the different notions of rationality developed in the field of cognitive science (i.e. classical rationality (CR), bounded rationality (BR) and Resource-Rationality (RR)) and will show their impact on the design and realization of AI systems. In particular, I will argue that, in order to build integrated AI systems able to exhibit a wide range of intelligent behaviors, it is of crucial importance to take into account bounded-rational and resource-rational cognitive constraints. In doing so, I will present two cognitively inspired AI applications - Dual PECCS and the TCL reasoning framework - showing how the outlined design perspective allowed such systems to 1) address some crucial aspects of commonsense reasoning in AI researh (namely: dealing with typicality effects and with the problem of commonsense compositionality) 2) integrate such systems with more general cognitive architectures 3) use their simulations as “computational explanations” to better understand the heuristics used by the human mind to face complex problems.

Download slide

Author’s short Bio

Antonio Lieto is a researcher in Computer Science at the Department of Computer Science of the University of Turin (Italy) and a Research Associate at the ICAR-CNR Institute in Palermo (Cognitive Systems for Robotics Lab). His research interests include computational models of cognition, commonsense reasoning, language and knowledge technologies. He is the current Vice-President of the Italian Association of Cognitive Sciences (AISC, 2017-2022), the recipient of the “Outstanding BICA Research Award” from the Biologically Inspired Cognitive Architecture Society (USA), and is an ACM Distinguished Speaker on the topics of cognitively inspired AI. He has authored the book “Cognitive Design for Artificial Minds” (Routledge/Taylor & Francis, 2021).