Advancing Cognitive Science and AI with Cognitive-AI Benchmarking

Catalog description

In recent years, the fields of cognitive science and artificial intelligence have converged in striking ways, impacting investigations of many aspects of human cognition, from perception and memory to language, problem solving, and social interaction. A key lesson from these developments is that continued progress will require systematic Cognitive-AI Benchmarking - the joint characterization of human behavior in a wide variety of naturalistic settings and rigorous testing of modern AI algorithms in those same settings. The purpose of this advanced graduate research course is to provide experienced researchers with a conceptual framework and set of tools to implement such coordinated investigations of human behavior and AI algorithms. Participants in this course will be expected to both discuss papers from the contemporary research literature in cognitively inspired AI, as well as make progress on their own Cognitive-AI Benchmarking project. This course is most likely to be appropriate for PhD students with substantial research experience in Psychology, Neuroscience, and/or AI.