Pattern languages: a new paradigm for neurocomputation

Andras, Peter

Abstract

Existing neurocomputation models do not describe very faithfully biological neural systems. Here we offer a new look at neural activity data, by introducing the paradigm of pattern languages. Biological examples are presented to provide the biological grounding, and an artificial neural network is described as a simple example of how to interpret neural activity data in terms of pattern languages. The new approach places the emphasis on internal processing within the neural system, instead of the input/output processing, and offers a good balance between the consideration of individual neural activity and of the activity of large numbers of neurons.

• Journal Article Type: Article
• Online Publication Date: Mar 2, 2004
• Publication Date: 2004-06
• Deposit Date: Nov 4, 2021
• Journal: Neurocomputing
• Print ISSN: 0925-2312
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 58-60
• Pages: 223-228
• DOI: https://doi.org/10.1016/j.neucom.2004.01.047
• Public URL: http://researchrepository.napier.ac.uk/Output/2809011