AN EFFICIENT MULTI-SPIN CODING ALGORITHM FOR NEURAL NETWORKS

作     者：Neschen, Martin 

作者机构：Univ Cologne Inst Theoret Phys D-5000 Cologne 41 Germany 

出 版 物：《INTERNATIONAL JOURNAL OF MODERN PHYSICS C》 (国际现代物理学杂志，C: 计算物理学和物理计算)

年 卷 期：1991年第2卷第2期

页      面：623-636页

核心收录：

学科分类：07[理学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 0702[理学-物理学] 

基　　金：Deutsche Forschungsgemeinschaft [Sonderforschungsbereich 341] 

主　　题：Neural Networks Multi-Spin Coding Computer Algorithms Large-Scale Simulations 

摘      要：A new method for large-scale numerical simulations of neural networks is proposed which reduces the computational effort by incrementally updating the local fields and thus restricting the operations to flipped spins only. A highly optimized multi-spin algorithm is described employing words oriented along the columns of the coupling matrix unlike the horizontal structure in existing high-speed algorithms. An effective rate of 35+109 couplings/s on a Cray-YMP can be attained which is about five times as fast as best existing multi-spin implementations.