Please use this identifier to cite or link to this item: http://hdl.handle.net/11400/10447
Title: Batch self-organizing map algorithm
Alternative title / Subtitle: a theoretical study of self-organization of a 1- D network under quantization effects
Authors: Βασιλάς, Νικόλαος
Item type: Conference publication
Conference Item Type: Short Paper
Keywords: αυτο-οργανούμενοι χάρτες;διαδικασίες Μάρκοβ;νευρωνικά δίκτυα;επιδράσεις της κβάντωσης;Self-organizing maps;Neural networks;quantization effects
Subjects: Πληροφορική
Μηχανική υπολογιστών
Computer science
Computer engineering
Issue Date: 15-May-2015
1-Jun-2008
Publisher: IEEE
Abstract: In this paper, we examine necessary and sufficient conditions that ensure self-organization of the batch variant of the self-organizing map algorithm for 1-D networks and for quantized weights and inputs. Using Markov chain formalism, it is shown that the existing analysis for the original algorithm can be extended to also include the more general batch variant. Finally, simulations verify the theoretical results, relate the speed of weight ordering to the distribution of the inputs and show the existence of metastable states of the Markov chain.
Language: English
Citation: Vassilas, N. (2008) Batch self-organizing map algorithm: A theoretical study of self-organization of a 1-D network under quantization effects. Proceedings IEEE International Joint Conference on Neural Networks (IJCNN'08), WCCI 2008. pp.1925-1930. Hong-Kong: IEEE.
Conference: IEEE International Joint Conference on Neural Networks (IJCNN'08), WCCI 2008
Access scheme: Embargo
License: Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ηνωμένες Πολιτείες
URI: http://hdl.handle.net/11400/10447
Appears in Collections:Δημοσιεύσεις

Files in This Item:
There are no files associated with this item.


This item is licensed under a Creative Commons License Creative Commons