Pregled bibliografske jedinice broj: 241929
New preprocessing methods for holographic neural networks
New preprocessing methods for holographic neural networks // Proceedings of the International Conference on Artificial Neural Nets and Genetic Algorithms (ANNGA '93) / Albrecht, R.F ; Reeves, C.R ; Steele, N.C. (ur.).
Beč: Springer, 1993. str. 190-197 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 241929 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
New preprocessing methods for holographic neural networks
Autori
Manger, Robert ; Souček, Branko
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the International Conference on Artificial Neural Nets and Genetic Algorithms (ANNGA '93)
/ Albrecht, R.F ; Reeves, C.R ; Steele, N.C. - Beč : Springer, 1993, 190-197
Skup
International Confeerence on Artificial Neural Nets and Genetic Algorithms (ANNGA '93)
Mjesto i datum
Innsbruck, Austrija, 14.04.1993. - 16.04.1993
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
holographic neural networks; data preprocessing; stimulus expansion; stimulus symmetrization
Sažetak
We propose two new methods for preprocessing of stimulus data in holographic neural networks. The first method achieves optimal data symmetrization, and it is based on estimating the actual distributions of elements within a stimulus vector. The second method serves for data expansion, and it relies on sine and cosine functions to produce dummy stimulus elements. We describe an implementation of our methods and report on some experiments. Our proposals improve the applicability of holographic networks. Namely, symmetrization assures accuracy in reproducing learned stimulus-response associations, while expansion increases learning capacity.
Izvorni jezik
Engleski
Znanstvena područja
Matematika, Računarstvo
POVEZANOST RADA
Ustanove:
Prirodoslovno-matematički fakultet, Matematički odjel, Zagreb
