Inference in probabilistic models with applications to communications and signal processing

Abstract

This talk shall briefly overview two inference techniques in machine learning, namely Gaussian Belief Propagation and Variational Inference, and demonstrates how they are applied to communications and signal processing problems. Surprisingly, these two machine learning inference techniques are general enough to tackle a wide range of problems. The applications to be covered include synchronization in large-scale networks, channel estimation under high mobility, power state estimation in smart grid, massive MIMO channel estimation, tensor decomposition, face classification, surveillance video objects separation, and image de-noising.