Real-Time on-line-learning support vector machine based on a fully-parallel analog VLSI processor

Authors:
Renyuan Zhang;Tadashi Shibata
Affiliations:
Department of Electrical Engineering and Information Systems, The University of Tokyo, Tokyo, Japan;Department of Electrical Engineering and Information Systems, The University of Tokyo, Tokyo, Japan
Venue:
ICAISC'12 Proceedings of the 11th international conference on Artificial Intelligence and Soft Computing - Volume Part II
Year:
2012

Citing 9
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
The Template Update Problem

IEEE Transactions on Pattern Analysis and Machine Intelligence
On-line ensemble SVM for robust object tracking

ACCV'07 Proceedings of the 8th Asian conference on Computer vision - Volume Part I
An on-chip-trainable Gaussian-Kernel analog support vector machine

IEEE Transactions on Circuits and Systems Part I: Regular Papers
Comparing support vector machines with Gaussian kernels to radialbasis function classifiers

IEEE Transactions on Signal Processing
A one-layer recurrent neural network for support vector machine learning

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
An image representation algorithm compatible with neural-associative-processor-based hardware recognition systems

IEEE Transactions on Neural Networks
Kerneltron: support vector "machine" in silicon

IEEE Transactions on Neural Networks
Analog neural network for support vector machine learning

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

An analog VLSI implementation of on-line learning Support Vector Machine (SVM) has been developed for the classification of high-dimensional pattern vectors. A fully-parallel self-learning circuitry employing analog high-dimensional Gaussian-generation circuits was used as an SVM processor. This SVM processor achieves a high learning speed (one clock cycle at 10 MHz) within compact chip area. Based on this SVM processor, an on-line learning system has been developed with the consideration of limited hardware resource. According to circuit simulation results, the image patterns from an actual database were all classified into correct classes by the proposed system. The ineffective samples are successfully identified in real-time and updated by on-line learning patterns.