CogSci Symposium: Generative and Discriminative Models in Cognitive Science
Brad Love and I are organizing a symposium at this year’s Cognitive Science Conference, with guest speakers Matt Jones and Tom Griffiths. Come check it out!
One popular distinction in machine learning is between discriminative and generative models (Ng & Jordan, 2001). Given the cross fertilization between research in human and machine learning, the time is ripe to ask whether the mind is a generative or discriminative learning device. This symposium tackles this question from a variety of perspectives. The aim is to explore the explanatory value of these two basic views of learning, which cut across existing distinctions in cognitive science (e.g., connectionist vs. Bayesian approaches).