Seminar: Victor V. Miagkikh, "Learning in Networks: from Spiking Neural Nets to Graphs" (5/4/2009)

Abstract: Hebbian learning is a know principal of unsupervised learning in networks. If two events happen "close in time" then the strength of connection between the network nodes producing those events increases. Is it a complete set of learning axioms? Given a reinforcement signal (reward) for a sequence of actions we can add another axiom: “reward controls plasticity”. Thus, we get a reinforcement learning algorithm that could be used for training spiking neural networks. The author will demonstrate the utility of this algorithm on a maze learning problem. These learning principles can be applied not only to neural, but also to other kinds of networks, such as economical influence networks for portfolio optimization and social networks for a movie recommendation engine.