Farmer, Sara-Jayne (1998). Computationally efficient natural gradient descent. Masters thesis, Aston University.
Abstract
This study examines the use of matrix momentum terms with the aim of creating a more computationally efficient natural gradient descent algorithm for on-line learning. It uses the statistical mechanics framework created by Saad and Solla to describe the evolution of order parameters for this algorithm in a two-layer student-teacher scenario, and compares this with results from matrix-momentum natural gradient learning of real datasets.
| Publication DOI: | https://doi.org/10.48780/publications.aston.ac.uk.00021456 |
|---|---|
| Additional Information: | Copyright © Sara-Jayne Farmer, 1998. Sara-Jayne Farmer asserts their moral right to be identified as the author of this thesis. This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognise that its copyright rests with its author and that no quotation from the thesis and no information derived from it may be published without appropriate permission or acknowledgement. If you have discovered material in Aston Publications Explorer which is unlawful e.g. breaches copyright, (either yours or that of a third party) or any other law, including but not limited to those relating to patent, trademark, confidentiality, data protection, obscenity, defamation, libel, then please read our Takedown Policy and contact the service immediately. |
| Institution: | Aston University |
| Last Modified: | 08 Apr 2025 15:52 |
| Date Deposited: | 19 Mar 2014 11:40 |
| Completed Date: | 1998 |
| Authors: |
Farmer, Sara-Jayne
|
![[img]](https://publications.aston.ac.uk/style/images/fileicons/text.png)