Class of '62 Auditorium (John Morgan Building), 3620 Hamilton Walk
Department of Neuroscience
University of Cambridge
Deep neural networks: a new framework for modelling brain information processing
Recent advances in neural network modelling have enabled major strides in computer vision and other artificial intelligence applications. Artificial neural networks are inspired by the brain and their computations could be implemented in biological neurons. Although designed with engineering goals, this technology provides the basis for tomorrow’s computational neuroscience. I will describe a framework for testing such models with massively multivariate brain-activity data. In order to compare representations between brains and models, we characterize the representational spaces by matrices of representational dissimilarities among stimuli. Deep convolutional neural nets trained for visual object recognition have internal representational spaces remarkably similar to those of the human and monkey ventral visual pathway. Modern neural net technology puts an expanding array of complex cognitive tasks within our computational reach. We are entering an exciting new era, in which we will be able to build neurobiologically faithful feedforward and recurrent computational models of how biological brains perform high-level feats of intelligence.