St Andrews Research Repository

St Andrews University Home
View Item 
  •   St Andrews Research Repository
  • Computer Science (School of)
  • Computer Science
  • Computer Science Theses
  • View Item
  •   St Andrews Research Repository
  • Computer Science (School of)
  • Computer Science
  • Computer Science Theses
  • View Item
  •   St Andrews Research Repository
  • Computer Science (School of)
  • Computer Science
  • Computer Science Theses
  • View Item
  • Login
JavaScript is disabled for your browser. Some features of this site may not work without it.

Using subgoal chaining to address the local minimum problem

Thumbnail
View/Open
JonathanPLewisPhDThesis.pdf (37.82Mb)
Date
2002
Author
Lewis, Jonathan Peter
Supervisor
Weir, Michael
Funder
Engineering and Physical Sciences Research Council (EPSRC)
Metadata
Show full item record
Altmetrics Handle Statistics
Abstract
A common problem in the area of non-linear function optimisation is that of not being able to guarantee finding the global optimum of the function in a feasible time especially when local optima exist. This problem applies to various areas of heuristic search. One of these areas concerns standard training techniques for feedforward neural networks. The element of heuristic search consists of attempting to find a neural weight state corresponding to the lowest training error. This problem may be termed the local minimum problem. The local minimum problem is addressed for feedforward neural networks. This is done by first establishing the conditions under which local minimum interference for the training process is to be expected. A target based approach to subgoal chaining in supervised learning is then investigated. This is a method to improve travel for neural networks by directing it more precisely through local subgoals than may be achieved through a more distant goal. It is shown however that linear subgoal chains are not sufficient to overcome the local minimum problem. Two novel training techniques are presented which use non-linear subgoal chains and are examined for their capability to address the local minimum problem. It is found that attempting to target a neural network to do something it cannot may lead to suboptimal training. It is also found that targeting a network to do something it is capable of generally leads to successful training. A novel system is presented which is designed to create optimal realisable targets for unrealisable goals. This allows neural networks to subsequently achieve the optimal weight state through a sufficiently powerful training method such as subgoal chaining. The results are shown to be consistent with the theoretical expectations.
Type
Thesis, PhD Doctor of Philosophy
Collections
  • Computer Science Theses
URI
http://hdl.handle.net/10023/14985

Items in the St Andrews Research Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

Advanced Search

Browse

All of RepositoryCommunities & CollectionsBy Issue DateNamesTitlesSubjectsClassificationTypeFunderThis CollectionBy Issue DateNamesTitlesSubjectsClassificationTypeFunder

My Account

Login

Open Access

To find out how you can benefit from open access to research, see our library web pages and Open Access blog. For open access help contact: openaccess@st-andrews.ac.uk.

Accessibility

Read our Accessibility statement.

How to submit research papers

The full text of research papers can be submitted to the repository via Pure, the University's research information system. For help see our guide: How to deposit in Pure.

Electronic thesis deposit

Help with deposit.

Repository help

For repository help contact: Digital-Repository@st-andrews.ac.uk.

Give Feedback

Cookie policy

This site may use cookies. Please see Terms and Conditions.

Usage statistics

COUNTER-compliant statistics on downloads from the repository are available from the IRUS-UK Service. Contact us for information.

© University of St Andrews Library

University of St Andrews is a charity registered in Scotland, No SC013532.

  • Facebook
  • Twitter