EE4540 Distributed signal processing

Topics: Signal processing techniques for decentralized signal processing
This is a course on decentralized signal processing techniques. Due to the explosion in size and complexity of modern datasets, it is increasingly important to be able to solve problems with a very large number of features or training examples. In industry, this trend has been referred to as ‘Big Data’, and it has had a significant impact in areas as varied as artificial intelligence, internet applications, computational biology, medicine, finance, marketing, journalism, network analysis, weather forecast, telecommunication, and logistics. As a result, both the decentralized collection or storage of these data sets as well as accompanying distributed solution methods are either necessary or at least highly desirable.

In this course we will focus on three signal processing techniques for decentralized processing: one based on distributed consensus, one on graphical models and one based on convex optimization. We will consider the following topics: synchronous and asynchronous versions of distributes averaging, gossip algorithms (randomized, geographic, broadcast, etc.), graphical models, probabilistic inference, message passing, min-sum/max-product algorithm, Jacobi and Gauss-Seidel algorithm, convex optimization, dual ascent, dual decomposition, alternating direction method of multipliers (ADMM) and primal-dual method of multipliers (PDMM).

Distributed consensus

Keywords: distributes averaging, synchronous vs. asynchronous processing, gossip, one-way averaging.

Graphical models

Keywords: Graphical models, trees, loopy graphs, conditional independence, Markov random fields, probabilisic inference, maximum a posteriori estimation, min-sum algorithm, linear coordinate descent algorithm, Jacobi algorithm, Gauss-Seidel algorithm, over-relaxation.

Convex optimization

Keywords: Convex optimization, convex sets, convex functions, dual function, KKT conditions, gossip algorithm, duals ascent algorithm, dual decomposition, augmented Lagrangian, alternating direction method of multipliers (ADMM), primal-dual method of multipliers (PDMM).

Teachers

dr.ir. Richard Heusdens

Audio and acoustic signal processing, distributed signal processing, information theory (source coding), speech enhancement

dr. Guoqiang Zhang

Audio signal processing, machine learning, distributed signal processing

Last modified: 2016-04-27

Details

Credits: 5 EC
Period: 0/0/4/0
Contact: Richard Heusdens