SVM learning with fixed-point math

We present in this paper an algorithm for Support Vector Machine (SVM) learning, which can be implemented using fixed-point math. The advantages of the fixed-point representation, respect to the more common floating-point one, allows us to address digital VLSI implementations of SVM. In particular, simple algorithms and simple architectures can be exploited for targeting programmable devices like Field Programmable Gate Arrays (FPGAs), which are the basis of many embedded systems. This paper focuses on the SVM learning algorithm: for the complete version of this work, including an actual FPGA realization.

[1]  Katya Scheinberg,et al.  Incremental Learning and Selective Sampling via Parametric Optimization Framework for SVM , 2001, NIPS.

[2]  Chih-Jen Lin,et al.  Asymptotic convergence of an SMO algorithm without any assumptions , 2002, IEEE Trans. Neural Networks.

[3]  Michel Renovell,et al.  Field-Programmable Logic and Applications: Reconfigurable Computing Is Going Mainstream , 2002, Lecture Notes in Computer Science.

[4]  C.F.N. Cowan,et al.  Adaptive equalization of finite nonlinear channels using multilayer perceptron , 1990 .

[5]  John C. Platt,et al.  Fast training of support vector machines using sequential minimal optimization, advances in kernel methods , 1999 .

[6]  Davide Anguita,et al.  Learning algorithm for nonlinear support vector machines suited for digital VLSI , 1999 .

[7]  S. Sathiya Keerthi,et al.  Convergence of a Generalized SMO Algorithm for SVM Classifier Design , 2002, Machine Learning.

[8]  Davide Anguita,et al.  Improved neural network for SVM learning , 2002, IEEE Trans. Neural Networks.

[9]  Pavel Laskov,et al.  Feasible Direction Decomposition Algorithms for Training Support Vector Machines , 2002, Machine Learning.

[10]  María José Pérez-Ilzarbe.  Convergence analysis of a discrete-time recurrent neural network to perform quadratic real optimization with bound constraints , 1998, IEEE Trans. Neural Networks.

[11]  Davide Anguita,et al.  A digital architecture for support vector machines: theory, algorithm, and FPGA implementation , 2003, IEEE Trans. Neural Networks.

[12]  Jun Wang,et al.  Global stability of a class of discrete-time recurrent neural networks , 2002 .

[13]  James A. Bucklew,et al.  Support vector machine techniques for nonlinear equalization , 2000, IEEE Trans. Signal Process..

[14]  Andrea Boni,et al.  SoftTOTEM: An FPGA Implementation of the TOTEM Parallel Processor , 2002, FPL.

[15]  Davide Anguita,et al.  Circuital implementation of support vector machines , 1998 .