Published November 2010
| Version v1
Book
The calculation of Moore-Penrose generalize inverse in OLAM neural network
Creators
- 1. Institute of Nuclear Science and Technology of Sichuan University, Chengdu (China)
Description
The Greville algorithm is the most common used method to calculate Moore-Penrose generalize inverse in OLAM (Optimal Linear Associative Memory) neural network. Both Greville and ordinary inverse method are used in the analysis of XRF spectra and compared in time and space complexity, some techniques must be used when implementing these two methods in computer are pointed out also. The results show that, the Greville algorithm can not meet the time and space requirement simultaneously when one of the row or column counts of the matrix is very large, but the ordinary in verse method is suitable when one of the row or column counts of the matrix is small. (authors)
Additional details
Publishing Information
- Publisher
- Atomic Energy Press
- Imprint Place
- Beijing (China)
- ISBN
- 978-7-5022-5040-9
- Imprint Title
- Progress report on nuclear science and technology in China (Vol.1). Proceedings of academic annual meeting of China Nuclear Society in 2009, No.9--nuclear technology applied in industry sub-volume
- Imprint Pagination
- 393 p.
- Journal Page Range
- p. 57-60
Conference
- Title
- academic annual meeting of China Nuclear Society
- Acronym
- '09
- Dates
- 18-20 Nov 2009
- Place
- Beijing (China)
INIS
- Country of Publication
- China
- Country of Input or Organization
- China
- INIS RN
- 45086446
- Subject category
- S97: MATHEMATICAL METHODS AND COMPUTING;
- Resource subtype / Literary indicator
- Conference
- Descriptors DEI
- ALGORITHMS; COMPARATIVE EVALUATIONS; ITERATIVE METHODS; MATRICES; NEURAL NETWORKS; X-RAY FLUORESCENCE ANALYSIS
- Descriptors DEC
- CALCULATION METHODS; CHEMICAL ANALYSIS; EVALUATION; MATHEMATICAL LOGIC; NONDESTRUCTIVE ANALYSIS; X-RAY EMISSION ANALYSIS
Optional Information
- Notes
- 1 tabs., 9 refs.