Advanced

Translating feed-forward nets to SOM-like maps

vanderZwaag, B J; Spaanenburg, Lambert LU and Slump, C (2003) 14th ProRISC Workshop on Circuits, Systems and Signal Processing, 2003 p.447-452
Abstract
A major disadvantage of feedforward neural

networks is still the difficulty to gain insight into their internal functionality. This is much less the case for, e.g., nets that are trained unsupervised, such as Kohonen’s self-organizing feature maps (SOMs). These offer a direct view into the stored knowledge, as their internal knowledge is stored in the same format as the input data that was used for training or is used for evaluation. This paper discusses a mathematical

transformation of a feed-forward network into a SOMlike

structure such that its internal knowledge can be visually

interpreted. This is particularly applicable to networks

trained in the general classification problem domain.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
feature maps, selforganizing maps, Neural networks, rule extraction, character recognition.
host publication
Proceedings ProRisc?03
pages
447 - 452
conference name
14th ProRISC Workshop on Circuits, Systems and Signal Processing, 2003
conference location
Veldhoven, Netherlands
conference dates
2003-11-26 - 2003-11-27
ISBN
90-73461-39-1
language
English
LU publication?
yes
id
2d468990-d93f-4b4c-b591-ce5fb5056bbb (old id 603852)
date added to LUP
2007-12-04 11:14:38
date last changed
2018-11-21 21:15:09
@inproceedings{2d468990-d93f-4b4c-b591-ce5fb5056bbb,
  abstract     = {A major disadvantage of feedforward neural<br/><br>
networks is still the difficulty to gain insight into their internal functionality. This is much less the case for, e.g., nets that are trained unsupervised, such as Kohonen’s self-organizing feature maps (SOMs). These offer a direct view into the stored knowledge, as their internal knowledge is stored in the same format as the input data that was used for training or is used for evaluation. This paper discusses a mathematical<br/><br>
transformation of a feed-forward network into a SOMlike<br/><br>
structure such that its internal knowledge can be visually<br/><br>
interpreted. This is particularly applicable to networks<br/><br>
trained in the general classification problem domain.},
  author       = {vanderZwaag, B J and Spaanenburg, Lambert and Slump, C},
  isbn         = {90-73461-39-1},
  keyword      = {feature maps,selforganizing
maps,Neural networks,rule extraction,character recognition.},
  language     = {eng},
  location     = {Veldhoven, Netherlands},
  pages        = {447--452},
  title        = {Translating feed-forward nets to SOM-like maps},
  year         = {2003},
}