Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Feed-forward support vector machine without multipliers.

Davide Anguita, Stefano Pischiutta, Sandro Ridella

IEEE Transactions on Neural Networks

|September 28, 2006

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Robustness-Congruent Adversarial Training for Secure Machine Learning Model Updates.

IEEE transactions on pattern analysis and machine intelligence·2025

Same author

Toward Learning Trustworthily from Data Combining Privacy, Fairness, and Explainability: An Application to Face Recognition.

Entropy (Basel, Switzerland)·2021

Same author

Distribution-Dependent Weighted Union Bound.

Entropy (Basel, Switzerland)·2021

Same author

Spectral Analysis of Electricity Demand Using Hilbert-Huang Transform.

Sensors (Basel, Switzerland)·2020

Same author

Learning With Kernels: A Local Rademacher Complexity-Based Analysis With Application to Graph Kernels.

IEEE transactions on neural networks and learning systems·2018

Same author

A local Vapnik-Chervonenkis complexity.

Neural networks : the official journal of the International Neural Network Society·2016

Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013

Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011

Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011

Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011

Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011

Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011

See all related articles

We introduce a novel algorithm for Support Vector Machine (SVM) computations using a Coordinate Rotation Digital Computer (CORDIC)-like approach. This method simplifies hardware implementation by using only shift and add operations, maintaining high classification accuracy.

Area of Science:

Machine Learning
Computer Engineering
Digital Signal Processing

Background:

Support Vector Machines (SVMs) are powerful classification algorithms, but their computational complexity, especially in the feed-forward phase, poses challenges for hardware implementation.
Conventional SVMs often rely on multiplications, which are resource-intensive in fixed-point arithmetic, limiting their applicability in embedded systems.

Discussion:

This work presents a Coordinate Rotation Digital Computer (CORDIC)-like algorithm tailored for the SVM feed-forward phase, specifically designed for fixed-point hardware.
The proposed algorithm exclusively utilizes shift and add operations, thereby eliminating the need for computationally expensive multipliers.
A novel hardware-friendly kernel is introduced, which simplifies the SVM computation while preserving classification efficacy.

Related Experiment Videos

Key Insights:

The CORDIC-like algorithm enables efficient SVM feed-forward computation using only shift and add operations.
The developed hardware-friendly kernel significantly reduces computational complexity.
The proposed approach achieves classification performance comparable to the conventional Gaussian kernel in SVMs.

Outlook:

This algorithm offers a promising solution for deploying SVMs on resource-constrained hardware platforms.
Further research could explore the application of this method to other machine learning algorithms with similar computational bottlenecks.
Optimization of the CORDIC-like approach for various fixed-point architectures could enhance its practical utility.