Peter Dayan

Fellow of the Royal Society Co-recipient of the Brain Prize in 2017

Peter Dayanstudied Mathematics at Cambridge University, did his PhD in ComputationalNeuroscience at the University of Edinburgh with David Willshaw, and postdocswith Terry Sejnowski at the Salk Institute and Geoff Hinton at the Universityof Toronto. After three years as an assistant professor at MIT, he helped foundthe Gatsby Computational Neuroscience Unit at UCL in 1998. He was Director ofthe Unit from 2002-2017, and is currently on sabbatical at Uber AI Labs. He wonthe Rumelhart Prize in 2012, and was a co-recipient of the Brain Prize in 2017.He has been recently elected as Fellow of the Royal Society in 2018.

 

Dayan's interestscentre on mathematical and computational models of neural processing, with aparticular emphasis on representation, learning and decision making. His recentwork has concentrated on the multifarious mechanisms by which humans and otheranimals approximately maximize reward and minimize punishment. The systemsinvolved apparently recapitulate algorithms popular in artificial intelligence;this connection offers a fecund route for understanding their normal behaviourand their characteristic dysfunctions in neurological and psychiatric disease.