Large Structures Seminar: Guido Montúfar

Given a body and an environment, what is the brain complexity needed in order to generate a desired set of behaviors? In this talk I will present a mathematical approach that links the physical and behavioral constraints of an agent to the required controller complexity. As controller architecture I consider a well-known artificial neural network, the conditional restricted Boltzmann machine, and define its complexity as the number of hidden units. I will also present experiments conducted with a virtual six-legged walking creature, which provide evidence for the accuracy of the theoretical predictions.