[{"publication_identifier":{"issn":["1370-4621"]},"type":"journal_article","first_author":"Heidemann, Gunther","keyword":["clustering","neural gas","codebook generation","vector quantization","competitive learning","unsupervised learning","winner takes all"],"doi":"10.1023/A:1009678928250","quality_controlled":"1","citation":{"frontiers":"Heidemann, G., and Ritter, H. (2001). Efficient vector quantization using the WTA-rule with activity equalization. *Neural Processing Letters* 13, 17-30.","mla":"Heidemann, Gunther, and Ritter, Helge. “Efficient vector quantization using the WTA-rule with activity equalization”. *Neural Processing Letters* 13.1 (2001): 17-30.","ieee":" G. Heidemann and H. Ritter, “Efficient vector quantization using the WTA-rule with activity equalization”, *Neural Processing Letters*, vol. 13, 2001, pp. 17-30.","angewandte-chemie":"G. Heidemann, and H. Ritter, “Efficient vector quantization using the WTA-rule with activity equalization”, *Neural Processing Letters*, **2001**, *13*, 17-30.","aps":" G. Heidemann and H. Ritter, Efficient vector quantization using the WTA-rule with activity equalization, Neural Processing Letters **13**, 17 (2001).","ama":"Heidemann G, Ritter H. Efficient vector quantization using the WTA-rule with activity equalization. *Neural Processing Letters*. 2001;13(1):17-30.","harvard1":"Heidemann, G., & Ritter, H., 2001. Efficient vector quantization using the WTA-rule with activity equalization. *Neural Processing Letters*, 13(1), p 17-30.","chicago":"Heidemann, Gunther, and Ritter, Helge. 2001. “Efficient vector quantization using the WTA-rule with activity equalization”. *Neural Processing Letters* 13 (1): 17-30.

","apa":"Heidemann, G., & Ritter, H. (2001). Efficient vector quantization using the WTA-rule with activity equalization. *Neural Processing Letters*, *13*(1), 17-30. doi:10.1023/A:1009678928250","apa_indent":"Heidemann, G., & Ritter, H. (2001). Efficient vector quantization using the WTA-rule with activity equalization. *Neural Processing Letters*, *13*(1), 17-30. doi:10.1023/A:1009678928250

","dgps":"Heidemann, G. & Ritter, H. (2001). Efficient vector quantization using the WTA-rule with activity equalization. *Neural Processing Letters*, *13*(1), 17-30. KLUWER ACADEMIC PUBL. doi:10.1023/A:1009678928250.

","bio1":"Heidemann G, Ritter H (2001)

Efficient vector quantization using the WTA-rule with activity equalization.

Neural Processing Letters 13(1): 17-30.","lncs":" Heidemann, G., Ritter, H.: Efficient vector quantization using the WTA-rule with activity equalization. Neural Processing Letters. 13, 17-30 (2001).","default":"Heidemann G, Ritter H (2001)

*Neural Processing Letters* 13(1): 17-30.","wels":"Heidemann, G.; Ritter, H. (2001): Efficient vector quantization using the WTA-rule with activity equalization *Neural Processing Letters*,13:(1): 17-30."},"publication":"Neural Processing Letters","intvolume":" 13","title":"Efficient vector quantization using the WTA-rule with activity equalization","year":"2001","_id":"1617662","id":"1617662","date_created":"2010-04-28T13:03:21Z","author":[{"last_name":"Heidemann","autoren_ansetzung":["Heidemann, Gunther","Heidemann","Gunther Heidemann","Heidemann, G","Heidemann, G.","G Heidemann","G. Heidemann"],"full_name":"Heidemann, Gunther","first_name":"Gunther"},{"autoren_ansetzung":["Ritter, Helge","Ritter","Helge Ritter","Ritter, H","Ritter, H.","H Ritter","H. Ritter"],"last_name":"Ritter","first_name":"Helge","full_name":"Ritter, Helge","id":"91130"}],"date_submitted":"2011-06-15T19:45:42Z","isi":"1","department":[{"_id":"10038"},{"_id":"10066"}],"language":[{"iso":"eng"}],"external_id":{"isi":["000167731600002"]},"abstract":[{"lang":"eng","text":"We propose a new algorithm for vector quantization, the Activity Equalization Vector quantization (AEV). It is based on the winner takes all rule with an additional supervision of the average node activities over a training interval and a subsequent re-positioning of those nodes with low average activities. The re-positioning is aimed to both an exploration of the data space and a better approximation of already discovered data clusters by an equalization of the node activities. We introduce a learning scheme for AEV which requires as previous knowledge about the data only their bounding box. Using an example of Martinetz et al. [1], AEV is compared with the Neural Gas, Frequency Sensitive Competitive Learning (FSCL) and other standard algorithms. It turns out to converge much faster and requires less computational effort."}],"status":"public","accept":"1","date_updated":"2018-07-24T12:59:31Z","issue":"1","publisher":"KLUWER ACADEMIC PUBL","article_type":"original","publication_status":"published","volume":"13","page":"17-30"}]