Title :
On a novel unsupervised competitive learning algorithm for scalar quantization
Author :
Van Hulle, Marc M. ; Martinez, Dominique
Author_Institution :
MIT, Cambridge, MA, USA
fDate :
5/1/1994 12:00:00 AM
Abstract :
This letter presents a novel unsupervised competitive learning rule called the boundary adaptation rule (BAR), for scalar quantization. It is shown both mathematically and by simulations that BAR converges to equiprobable quantizations of univariate probability density functions and that, in this way, it outperforms other unsupervised competitive learning rules
Keywords :
neural nets; probability; unsupervised learning; boundary adaptation rule; equiprobable quantizations; scalar quantization; univariate probability density functions; unsupervised competitive learning algorithm; Data communication; Image converters; Image recognition; Nearest neighbor searches; Neurons; Probability density function; Quantization; Speech coding; Speech recognition; Statistical analysis;
Journal_Title :
Neural Networks, IEEE Transactions on
