Parallel Factorised Algorithms for Mixture Estimation
This paper describes software aspects of an advisory system based on finite-mixture estimation. Factorised algorithms have been designed. Parallelism is used as a principal approach to acceleration of learning and processing phases. A coarse grain granularity parallelism is used in the first phase of the work. The Parallel Virtual Machine (PVM) is used for parallel implementation.
KeywordsSequential Algorithm Recursive Little Square Factorise Algorithm Recursive Little Square Algorithm Parallel Virtual Machine
Unable to display preview. Download preview PDF.
- M. Kárný et al, “ProDaCTool — theory, algorithms, and software”, EU Project IST ProDaC Tool No. IST-1999-12058, Research Report, 2000.Google Scholar
- M. Moonen, “Introduction to Adaptive Signal Processing”, K.U. Leuven, Leuven, Belgium, 1999.Google Scholar
- P. Nedoma et al, “MixTools, MATLAB Toolbox for Mixtures”, EU Project IST ProDaCTool No. IST-1999-12058, Research Report, 2000.Google Scholar
- P. Tvrdik, “Parallel Systems and Algorithms”, Lecture Notes, Czech Technical University, August 1996.Google Scholar