DATE: Wednesday May 9th, 2:10pm
LOCATION: MSB 1147, Colloquium Room
REFRESHMENTS: 3:30pm, MSB 4110
SPEAKER: David Schwab, Department of Biology and Physics, City University of New York
TITLE: “Perspectives from physics on learning and inference ”
ABSTRACT: I will give a physics perspective to deep learning, a popular set of techniques in machine learning where performance on tasks such as visual object recognition rivals human performance. I present work relating greedy training of so-called deep belief networks to variational real-space renormalization, a method from physics for simplifying complex systems. This connection may help explain how deep networks automatically learn relevant features from data as well as for what types of data deep networks work best. I will then discuss work using quantum-inspired tensor networks for supervised learning. Tensor networks are efficient representations of high-dimensional tensors that have been very successful in modeling many-body physics systems. Methods for optimizing tensor networks can be adapted to learning problems, and we find good performance on classic datasets. I will speculate on why this method works, using a perspective from physics that suggests a natural way forward.