Preparing for knowledge extraction in modular neural networks
(2002) 3rd IEEE Benelux Signal Processing Symposium (SPS) p.121-124- Abstract
- Neural networks learn knowledge from data. For a monolithic structure, this knowledge can be easily used but not isolated. The many degrees of freedom while learning make knowledge extraction a computationally intensive process as the representation is not unique. Where existing knowledge is inserted to initialize the network for training, the effect becomes subsequently randomized within the solution space. The paper describes structuring techniques such as modularity and hierarchy to create a topology that provides a better view on the learned knowledge to support a later rule extraction.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/603921
- author
- Spaanenburg, Lambert LU ; Slump, C H ; Venema, R S and vanderZwaag, B J
- publishing date
- 2002
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- 3rd IEEE Signal Processing Symposium SPS'02
- pages
- 121 - 124
- conference name
- 3rd IEEE Benelux Signal Processing Symposium (SPS)
- conference location
- Leuven, Belgium
- conference dates
- 2002-03-21 - 2002-03-22
- language
- English
- LU publication?
- no
- id
- fb954244-adb4-431a-879e-52a228a4dea8 (old id 603921)
- date added to LUP
- 2016-04-04 13:45:29
- date last changed
- 2018-11-21 21:16:07
@inproceedings{fb954244-adb4-431a-879e-52a228a4dea8, abstract = {{Neural networks learn knowledge from data. For a monolithic structure, this knowledge can be easily used but not isolated. The many degrees of freedom while learning make knowledge extraction a computationally intensive process as the representation is not unique. Where existing knowledge is inserted to initialize the network for training, the effect becomes subsequently randomized within the solution space. The paper describes structuring techniques such as modularity and hierarchy to create a topology that provides a better view on the learned knowledge to support a later rule extraction.}}, author = {{Spaanenburg, Lambert and Slump, C H and Venema, R S and vanderZwaag, B J}}, booktitle = {{3rd IEEE Signal Processing Symposium SPS'02}}, language = {{eng}}, pages = {{121--124}}, title = {{Preparing for knowledge extraction in modular neural networks}}, year = {{2002}}, }