Christopher Summerfield

Learning and generalisation of task knowledge in humans and neural networks

There has been a renaissance of interest in connectionist networks as models of biological computation. During sensory perception, deep neural networks learn representations that resemble those in primate neocortex. However, neural networks learn to perform and generalise cognitive tasks in very different ways to people. In my talk, I will explore these differences, and suggest computational adaptations that allow neural networks to learn multiple tasks in series, reconfigure task knowledge from limited data, and generalise knowledge between tasks.

Christopher Summerfield is Professor of Cognitive Neuroscience at the University of Oxford. His research group studies how humans learn and make decisions, using computer-based tasks, computational modelling and functional neuroimaging. Recent projects are focussed on how humans acquire new concepts or patterns in data, and how they use this information to make decisions in novel settings.  He also works at the AI research company Deepmind.