On November 18th, Professor Michael J. Tarr, Kavčić-Moura University Professor of Cognitive and Brain Science of the Department of Psychology at Carnegie Mellon University’s Dietrich College of Humanities and Social Sciences and at the Neuroscience Institute gave a seminar at INESC ID gathering faculty and researchers interested in neuroscience, cognitive sciences, and artificial intelligence.
Titled “Inductive Biases in Artificial Systems”, this seminar explored how insights from biological systems, particularly the brain, can inform the design of more effective artificial intelligence.
Professor Tarr emphasized the importance of leveraging knowledge of the visual systems of primates and other animals to inspire AI models that learn more efficiently, operate with greater robustness, and consume less energy. He highlighted how these biological principles could guide the development of vision systems capable of interpreting data with reduced input requirements.
The seminar took place during his sabbatical in Portugal, where he is collaborating with researchers from Instituto Superior Técnico, Universidade de Coimbra, and the Champalimaud Foundation.