The effect of the coefficients used in the conventional back propagation algorithm on training connectionist models is discussed, using a vowel recognition task in speech processing as an example. Some weaknesses of the use of fixed coefficients are described and an adaptive algorithm using variable
โฆ LIBER โฆ
An adaptive parallel algorithm for analyzing activity networks
โ Scribed by Pranay Chaudhuri
- Publisher
- Elsevier Science
- Year
- 1990
- Tongue
- English
- Weight
- 272 KB
- Volume
- 9
- Category
- Article
- ISSN
- 0167-6377
No coin nor oath required. For personal study only.
๐ SIMILAR VOLUMES
An adaptive training algorithm for back
โ
L.-W. Chan; F. Fallside
๐
Article
๐
1987
๐
Elsevier Science
๐
English
โ 628 KB
An efficient parallel algorithm for LISS
โ
Li-Chiu Chang; Fi-John Chang
๐
Article
๐
2002
๐
Elsevier Science
๐
English
โ 805 KB
An adaptive parallel genetic algorithm s
โ
Yuk-Yin Wong; Kin-Hong Lee; Kwong-Sak Leung
๐
Article
๐
2003
๐
John Wiley and Sons
๐
English
โ 501 KB
A parallel FFT algorithm for transputer
โ
Yonggen Huang; Yakup Paker
๐
Article
๐
1991
๐
Elsevier Science
๐
English
โ 468 KB
Huang, Y. and Y Paker, A parallel FFT algorithm for transputer networks, Parallel Computing 17 (1991) 895-906 This paper presents a parallel FFT algorithm suitable for MIMD machines\_ It has the properties of uniform granularity, the dominance of computation, and suitability to large data matrices T
An adaptive quality of service channel b
โ
Ibrahim Habib; Mahmoud Sherif; Mahmoud Naghshineh; Parviz Kermani
๐
Article
๐
2003
๐
John Wiley and Sons
๐
English
โ 209 KB
๐ 1 views
LLACA: An adaptive localized clustering
โ
Javad Akbari Torkestani; Mohammad Reza Meybodi
๐
Article
๐
2011
๐
Elsevier Science
๐
English
โ 896 KB