Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.3/44068

Download PDF (Published version) (Adobe Acrobat PDF, -1 bytes)

Title

Comments on 'On a novel unsupervised competitive learning algorithm for scalar quantization'

Author(s)

Andrew, Lachlan L. H.

Abstract

A recent letter presented a novel neural-network learning rule, BAR, (boundary adaptation rule) which was shown to converge to a scalar quantizer with equiprobable outputs. Such quantizers will be called maximum entropy quantizers (MEQs). It is interesting that such a simple rule can produce these quantizers. Its practical usefulness is limited, however, by two factors. First, there are more efficient algorithms which yield better results, and second MEQs are unsuitable for many quantization tasks, as discussed below.

Publication type

Journal article

Source

IEEE Transactions on Neural Networks, Vol. 7, no. 1 (Jan 1996), pp. 254-256

Publication year

1996

Keyword(s)

BAR;  Boundary adaptation rule;  Learning algorithms;  Learning systems;  Maximum entropy quantizers;  MEQs;  Neural networks;  Novel unsupervised competitive learning algorithm;  Performance;  Scalar quantisation;  Vector quantisation

Publisher

IEEE

ISSN

1045-9227

Publisher URL

http://dx.doi.org/10.1109/72.478412

Copyright

Copyright © 1996 IEEE. Paper reproduced here in accordance with the copyright policy of the publisher.

Additional information

This work was supported by a scholarship from the Australian Telecommunications and Electronics Research Board (ATERB).

Full text

Peer reviewed