A support vector machine with integer parameters

Authors:
Davide Anguita;Alessandro Ghio;Stefano Pischiutta;Sandro Ridella
Affiliations:
DIBE - University of Genoa, Via Opera Pia 11A, 16145 Genoa, Italy;DIBE - University of Genoa, Via Opera Pia 11A, 16145 Genoa, Italy;DIBE - University of Genoa, Via Opera Pia 11A, 16145 Genoa, Italy;DIBE - University of Genoa, Via Opera Pia 11A, 16145 Genoa, Italy
Venue:
Neurocomputing
Year:
2008

Citing 21
Cited 2

Discrete-time signal processing

Discrete-time signal processing
Using simulations of reduced precision arithmetic to design a neuro-microprocessor

Journal of VLSI Signal Processing Systems - Special issue on VLSI neural networks
Support-Vector Networks

Machine Learning
Mixing floating- and fixed-point formats for neural network learning on neuroprocessors

Microprocessing and Microprogramming
Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
Computer arithmetic: algorithms and hardware designs

Computer arithmetic: algorithms and hardware designs
Soft Margins for AdaBoost

Machine Learning
Numerical Experience with Lower Bounds for MIQP Branch-And-Bound

SIAM Journal on Optimization
Convergence of a Generalized SMO Algorithm for SVM Classifier Design

Machine Learning
The case for multi--tier camera sensor networks

NOSSDAV '05 Proceedings of the international workshop on Network and operating systems support for digital audio and video
Convergent Lagrangian and Contour Cut Method for Nonlinear Integer Programming with a Quadratic Objective Function

SIAM Journal on Optimization
An Experimental Study on Pedestrian Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
QP Algorithms with Guaranteed Accuracy and Run Time for Support Vector Machines

The Journal of Machine Learning Research
Dynamic reconfiguration in sensor networks with regenerative energy sources

Proceedings of the conference on Design, automation and test in Europe
Array processing for target DOA, localization, and classification based on AML and SVM algorithms in sensor networks

IPSN'03 Proceedings of the 2nd international conference on Information processing in sensor networks
Nonparametric decentralized detection using kernel methods

IEEE Transactions on Signal Processing
Successive overrelaxation for support vector machines

IEEE Transactions on Neural Networks
Asymptotic convergence of an SMO algorithm without any assumptions

IEEE Transactions on Neural Networks
A digital architecture for support vector machines: theory, algorithm, and FPGA implementation

IEEE Transactions on Neural Networks
Feed-Forward Support Vector Machine Without Multipliers

IEEE Transactions on Neural Networks
The design of a neuro-microprocessor

IEEE Transactions on Neural Networks

Electricity price forecasting based on support vector machine trained by genetic algorithm

IITA'09 Proceedings of the 3rd international conference on Intelligent information technology application
Applicability of approximate multipliers in hardware neural networks

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

We describe here a method for building a support vector machine (SVM) with integer parameters. Our method is based on a branch-and-bound procedure, derived from modern mixed integer quadratic programming solvers, and is useful for implementing the feed-forward phase of the SVM in fixed-point arithmetic. This allows the implementation of the SVM algorithm on resource-limited hardware like, for example, computing devices used for building sensor networks, where floating-point units are rarely available. The experimental results on well-known benchmarking data sets and a real-world people-detection application show the effectiveness of our approach.