Repository logo
 

Distributed associative memory


No Thumbnail Available

Type

Thesis

Change log

Authors

Sterne, Philip Jonathan 

Abstract

This dissertation modifies error-correcting codes and Bloom filters to create high-capacity associative memories. These associative memories use principled statistical inference and are distributed as no single component contains enough information to complete the task by itself, yet the components can collectively solve the task by passing information to each other. These associative memories are also robust to hardware failure as their distributed nature ensures there is no single point of failure. This dissertation starts by simplifying a Bloom filter so that it tolerates hardware failure (albeit with reduced performance). An efficient associative memory is created by performing inference over the set of items stored in the Bloom filter. This architecture suggests a modification which forgets old patterns stored in the associative memory (known as a palimpsest memory). It is shown that overwriting old patterns in an independent manner reduces performance, but is still comparable to the well-known Hopfield network. The lost performance can be regained using integer storage which allows the superposition of the pattern representation, or ensuring bits are not overwritten independently using concepts from errorcorrecting codes. The final task performs recall in continuous time using components which are more similar to neurons than used in the rest of the dissertation. The resulting memory has the exciting ability to recall many patterns simultaneously. Statistical inference ensures gradual degradation of the performance as an associative memory is overloaded. Since many definitions of associative memory capacity rely on the existence of catastrophic failure a new definition of capacity is provided. In spite of some biologically unrealistic attributes, this work is relevant to the understanding of the brain as it provides high performance solutions to the associative memory task which is known to be relevant to the brain.

Description

This thesis is not available on this repository until the author agrees to make it public. If you are the author of this thesis and would like to make your work openly available, please contact us: thesis@repository.cam.ac.uk.


Cambridge University Library can make a copy of this work available only for the purposes of private study and non-commercial research. Copies should not be shared or saved in any shared facilities. Copyright over the content of these works is with their authors. Theses from the Library collection are considered unpublished works and according to UK legislation quoting from them is not allowed without permission from their author.

If you can commit to these terms, please complete the request form which you can find through this link: https://imagingservices.lib.cam.ac.uk/


Please note that print copies of theses may be available for consultation in the Cambridge University Library's Manuscript reading room. Admission details are at http://www.lib.cam.ac.uk/collections/departments/manuscripts-university-archives

Date

Advisors

Keywords

Qualification

Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge