A dynamic system framework for the decomposition method solving Support Vector Machines

Full text for this resource is not available from the Research Repository.

Lai, Daniel ORCID: 0000-0003-3459-7709, Mani, N and Palaniswami, M (2004) A dynamic system framework for the decomposition method solving Support Vector Machines. In: ‬Proceedings of the 2004 Intelligent Sensors, Sensor Networks & Information Processing Conference :‎ ‬14-17 December 2004, Melbourne, Australia. IEEE, Adelaide, South Australia, pp. 283-288.

Abstract

The decomposition method is generally used to solve the quadratic program of Support Vector Machines. The rate of convergence of this method is largely dependant on the sequence of sub-problems solved. In order to study ways of increasing the convergence, we propose a dynamic system perspective to model the dynamics of the decomposition method. In particular, the minimization of a sub-problem can be viewed as an autonomous dissipative system in terms of second order differential equations. The gradients of the sub-problems and the inequality constraints are explicitly modelled as system variables. Using these models, we then define a general decomposition method as a non-autonomous system composed of sub-systems that operate for discrete time intervals. The dependance of this system on time is depicted by a time dependant permutation matrix which functions as an indicator for operating subsystem components.

Additional Information

‬IEEE cat. no. 04EX994C
‬Alternative titles :
ISSNIP 2004‬
Intelligent Sensors, Sensor Networks and Information Processing Conference

Item type Book Section
URI https://vuir.vu.edu.au/id/eprint/5558
Official URL https://ieeexplore.ieee.org/document/1417476
ISBN 0780388941
Subjects Historical > FOR Classification > 0102 Applied Mathematics
Historical > FOR Classification > 0802 Computation Theory and Mathematics
Historical > FOR Classification > 0902 Automotive Engineering
Historical > Faculty/School/Research Centre/Department > Centre for Ageing, Rehabilitation, Exercise & Sport Science (CARES)
Keywords ResPubID19078, support vector machines, SVM, nonlinear function, decomposition method, computing memory, optimization, chunking, algorithm, working set, search direction, a linear convergence rate, convergence
Citations in Scopus 0 - View on Scopus
Download/View statistics View download statistics for this item

Search Google Scholar

Repository staff login