×

Magnification control in self-organizing maps and neural gas. (English) Zbl 1110.68134

Summary: We consider different ways to control the magnification in Self-Organizing Maps (SOM) and Neural Gas (NG). Starting from early approaches of magnification control in vector quantization, we then concentrate on different approaches for SOM and NG. We show that three structurally similar approaches can be applied to both algorithms that are localized learning, concave-convex learning, and winner-relaxing learning. Thereby, the approach of concave-convex learning in SOM is extended to a more general description, whereas the concave-convex learning for NG is new. In general, the control mechanisms generate only slightly different behavior comparing both neural algorithms. However, we emphasize that the NG results are valid for any data dimension, whereas in the SOM case, the results hold only for the one-dimensional case.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDF BibTeX XML Cite
Full Text: DOI arXiv Link

References:

[1] DOI: 10.1016/0893-6080(90)90071-R
[2] DOI: 10.1007/BF02460791 · Zbl 0477.92008
[3] DOI: 10.1162/neco.1996.8.4.757 · Zbl 05477964
[4] DOI: 10.1016/S0893-6080(99)00027-1
[5] DOI: 10.1109/72.143371
[6] DOI: 10.1162/089976698300017953
[7] Brause R., Int. J. Computers and Artificial Intelligence 11 pp 173– (1992)
[8] DOI: 10.1109/34.682189
[9] DOI: 10.1023/A:1011326007550 · Zbl 0981.68699
[10] DOI: 10.1002/cplx.10084
[11] DOI: 10.1162/0899766053491922 · Zbl 1096.68129
[12] DOI: 10.1016/j.neucom.2004.01.191 · Zbl 02223919
[13] DOI: 10.1016/S0925-2312(98)00034-4 · Zbl 0917.68082
[14] DOI: 10.1016/j.neucom.2003.09.009 · Zbl 02060562
[15] DOI: 10.1109/72.363433
[16] DOI: 10.1038/326689a0
[17] DOI: 10.1016/0167-2789(92)90023-G · Zbl 0759.58030
[18] DOI: 10.1007/BF00201801 · Zbl 0747.92006
[19] DOI: 10.1109/72.501731
[20] DOI: 10.1016/0167-2789(83)90298-1 · Zbl 0593.58024
[21] DOI: 10.1162/089976699300016098
[22] DOI: 10.3758/BF03212211
[23] DOI: 10.1126/science.1736364
[24] DOI: 10.1109/TCOM.1980.1094577
[25] DOI: 10.1162/neco.1989.1.3.402
[26] DOI: 10.1109/72.88162
[27] DOI: 10.1109/72.238311
[28] DOI: 10.1016/0893-6080(94)90109-0 · Zbl 00695981
[29] DOI: 10.1109/72.80310
[30] DOI: 10.1007/BF00320480 · Zbl 0586.92004
[31] DOI: 10.1103/PhysRevA.41.3038
[32] Villmann Th., Neural Network World 10 pp 739– (2000)
[33] DOI: 10.1016/S0925-2312(01)00640-3 · Zbl 1006.68781
[34] Villmann Th., Neural Network World 10 pp 751– (2000)
[35] DOI: 10.1016/S0893-6080(03)00021-2 · Zbl 02022308
[36] DOI: 10.1098/rspb.1976.0087
[37] DOI: 10.1109/TIT.1982.1056490 · Zbl 0476.94008
[38] DOI: 10.1109/72.478394
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.