This paper presents a mapping scheme for the proposed implementation of neural network models on systolic arrays. The mapping technique is illustrated on the multilayer perceptron with back-propagation learning. Dependency graphs have been given that represent the operations in the execution phases
β¦ LIBER β¦
Mapping dynamic programming onto a linear systolic array
β Scribed by Prasanna Kumar, V. K. ;Tsai, Yu-Chen
- Book ID
- 105008290
- Publisher
- Springer
- Year
- 1990
- Tongue
- English
- Weight
- 684 KB
- Volume
- 1
- Category
- Article
- ISSN
- 0922-5773
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
Mapping of Neural Network Models onto Sy
β
Sudipta Mahapatra; Rabi N. Mahapatra
π
Article
π
2000
π
Elsevier Science
π
English
β 220 KB
Mapping matrix multiplication algorithm
β
N.M. StojanoviΔ; E.I. MilovanoviΔ; I. StojmenoviΔ; T.. MilovanoviΔ; T.I. TokiΔ
π
Article
π
2004
π
Elsevier Science
π
English
β 891 KB
An approach to design fault-tolerant hexagonal systolic array (SA) for multiplication of rectangular matrices is described. The approach comprises three steps. First, redundancies are introduced at the computational level by deriving three equivalent algorithms but with disjoint index spaces. Second
Mapping 3-D IIR digital filter onto syst
β
F. El-Guibaly; A. Tawfik
π
Article
π
1996
π
Springer US
π
English
β 988 KB
A systolic algorithm for dynamic program
β
Chau-Jy Lin
π
Article
π
1994
π
Elsevier Science
π
English
β 710 KB
Part II A systolic array for numerical l
π
Article
π
1993
π
Springer-Verlag
β 363 KB
A notational approach to formulation of
β
M.P. Bekakos
π
Article
π
1995
π
Elsevier Science
π
English
β 420 KB