Title :
Feed-Forward Support Vector Machine Without Multipliers
Author :
Anguita, Davide ; Pischiutta, S. ; Ridella, S. ; Sterpi, D.
Author_Institution :
Dept. of Biophys. & Electron. Eng., Genoa Univ.
Abstract :
In this letter, we propose a coordinate rotation digital computer (CORDIC)-like algorithm for computing the feed-forward phase of a support vector machine (SVM) in fixed-point arithmetic, using only shift and add operations and avoiding resource-consuming multiplications. This result is obtained thanks to a hardware-friendly kernel, which greatly simplifies the SVM feed-forward phase computation and, at the same time, maintains good classification performance respect to the conventional Gaussian kernel
Keywords :
feedforward neural nets; fixed point arithmetic; support vector machines; conventional Gaussian kernel; coordinate rotation digital computer; feedforward phase computation; feedforward support vector machine; fixed-point arithmetic; Algorithm design and analysis; Embedded computing; Embedded system; Feedforward systems; Fixed-point arithmetic; Hardware; Kernel; Machine learning algorithms; Support vector machine classification; Support vector machines; Coordinate rotation digital computer (CORDIC); embedded systems; fixed-point arithmetic; support vector machine (SVM); Algorithms; Artificial Intelligence; Cluster Analysis; Computing Methodologies; Neural Networks (Computer); Pattern Recognition, Automated;
Journal_Title :
Neural Networks, IEEE Transactions on
DOI :
10.1109/TNN.2006.877537
