Description

Ph.D. Studentship At The University of Surrey is going to be offered for the project named, “Information-theoretic learning for sound analysis“.

The purpose of this Ph.D. thesis is to look at informational theoretic methods for sound analysis. The Information Bottleneck (IB) approach has become a popular way to study learning in deep learning networks and autoencoders. The Information Bottleneck method works with Renyi and other entropies as well as standard Shannon entropy. Information estimate that is both quick and accurate is still a hot topic of study. This research will look at information-theoretic techniques for analyzing sound sequences, both for supervised and unsupervised learning methods like convolutive and recurrent networks and variational autoencoders. The research will also look at direct information loss estimators and new information-theoretic processing structures for sound processing, such as feed-forward and feedback connections inspired by biological brain networks’ transfer information. Ph.D. Studentship

Professor Mark Plumbley will supervise the research as part of the EPSRC Fellowship on AI for Sound.

Eligibility for Ph.D. Studentship At The University of Surrey

• Students must have a first-class degree
• Strong research interest
• IELTS above 6.5

How To Apply for Ph.D. Studentship

• All applications should be sent via Vision, Speech, and Signal Processing Ph.D. on the Apply Tab.
• Clearly mention Studentship Project in the subject line.
• For more information contact  [email protected] 
• Further Info on How To Apply Click Here

Benefits

For the academic year 21/22, a stipend of £15,609 will be paid, with the amount increasing each year in line with the UK Research and Innovation (UKRI) rate, and a £4,500 Home/EU-rate fee allowance (with automatic increase to UKRI rate each year). The scholarship is for three years. The University of Surrey provided funding for this project.