Cart
Using Relevance to Reduce Network Size Automatically
- DOI:
- 10.1080/09540098908915626
pages 3-16
Available online: 05 Apr 2007Abstract
This paper proposes a means of using the knowledge in a network to determine the functionality or relevance of individual units, both for the purpose of understanding the network's behavior and improving its performance. The basic idea is to iteratively train the network to a certain performance criterion, compute a measure of relevance that identifies which input or hidden units are most critical to performance, and automatically remove the least relevant units. This skeletonization technique can be used to simplify networks by eliminating units that convey redundant information; to improve learning performance by first learning with spare hidden units and then removing the unnecessary ones, thereby constraining generalization; and to understand the behavior of networks in terms of minimal ‘rules’.
- Recommend to:
- A friend
First page preview
Details
- Available online: 05 Apr 2007
Author affiliations
- a Department of Computer Science &, Institute of Cognitive Science, University of Colorado, Boulder, CO, 80309-0430, USA.
Librarians
- Librarians' area
- Pricing
- Institutional account
- Access entitlements
- Co-branding
- IP ranges
- Link resolver preferences
- Usage reports