An EPSRC-funded Ph.D studentship is available for the following project (for UK/EU students only). The studentship is a full award (fees+stipend) if the student is from the UK, and covers fees if the student is from the EU.
Project Title: Message-Passing Algorithms for Sparse Approximation and Data Compression
Description: Message passing algorithms and their approximations have been used for probabilistic inference in a variety of applications, e.g., for fast decoding of error-correcting codes, and for sparse signal recovery.
This project will explore the use of message-passing algorithms for finding good sparse approximations of high-dimensional signals. In the simplest setting, the goal is to represent a vector by efficiently finding the best sparse linear combination of vectors from a given dictionary. The problem of sparse approximation is closely related to lossy data compression. The algorithms developed in the will be used to compress continuous-valued sources with a sparse regression codebook.
Candidate Profile: An undergraduate degree in Engineering, Applied Mathematics, or Statistics with a good academic record. A strong mathematical background is essential, especially in probability and inference. Some knowledge of optimization and/or information theory is desirable. Prior research experience in any of these areas would be a plus.