Speaker
Tom Griffiths
Bio
Tom Griffiths is Henry R. Luce Professor of Information Technology, Consciousness, and Culture, at the Department of Psychology and Computer Science at Princeton University. He is also the director of Princeton AI Lab, a new effort that supports innovative research efforts in AI and related fields. His research focuses on developing mathematical models of human cognition, emphasizing what makes human intelligence distinct from modern AI. By studying how humans operate under constraints like limited time, computational capacity, and communication bandwidth, he seeks to understand how we achieve efficient learning, resourceful thinking, and collaboration. These insights aim to inform the design of AI systems that better mirror human-like intelligence.
Abstract
Psychologists and computer scientists have very different views of the mind. Psychologists tell us that humans are error-prone, using simple heuristics that result in systematic biases. Computer scientists view human intelligence as aspirational, trying to capture it in artificial intelligence systems. How can we reconcile these two perspectives? In this talk, I will argue that we can do so by reconsidering how we think about rational action. Rather than adopting a traditional standard for rationality, which focuses on choosing the best action without considering the computational difficulty of that choice, we can adopt a standard inspired by computer science in which the quality of the outcome trades off with the amount of computation involved. As a result, we obtain new models of human behavior that can help us understand the cognitive strategies that people adopt. I will present examples of this approach in the context of human decision-making and planning, including complex planning problems such as the game of chess.