A support vector machine with integer parameters

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.

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

[2]  D. Anguita,et al.  The effect of quantization on support vector machines with Gaussian kernel , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

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

[4]  Tomaso A. Poggio,et al.  People recognition and pose estimation in image sequences , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[5]  Gunnar Rätsch,et al.  Soft Margins for AdaBoost , 2001, Machine Learning.

[6]  R. J. Dakin,et al.  A tree-search algorithm for mixed integer programming problems , 1965, Comput. J..

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

[8]  Fu Lee Wang,et al.  Convergent Lagrangian and Contour Cut Method for Nonlinear Integer Programming with a Quadratic Objective Function , 2006, SIAM J. Optim..

[9]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[10]  John Wawrzynek,et al.  Using simulations of reduced precision arithmetic to design a neuro-microprocessor , 1993, J. VLSI Signal Process..

[11]  Stephen A. Dyer,et al.  Digital signal processing , 2018, 8th International Multitopic Conference, 2004. Proceedings of INMIC 2004..

[12]  Amos Omondi.  Neurocomputers: A Dead End? , 2000, Int. J. Neural Syst..

[13]  Dariu Gavrila,et al.  An Experimental Study on Pedestrian Classification , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Davide Anguita,et al.  Theoretical and Practical Model Selection Methods for Support Vector Classifiers , 2004 .

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

[16]  Kung Yao,et al.  Array Processing for Target DOA, Localization, and Classification Based on AML and SVM Algorithms in Sensor Networks , 2003, IPSN.

[17]  Prashant J. Shenoy,et al.  The case for multi--tier camera sensor networks , 2005, NOSSDAV '05.

[18]  Mahesan Niranjan,et al.  Uncertainty in geometric computations , 2002 .

[19]  John Wawrzynek,et al.  The design of a neuro-microprocessor , 1993, IEEE Trans. Neural Networks.

[20]  Ingo Steinwart,et al.  Approximate Duality , 2007 .

[21]  Behrooz Parhami,et al.  Computer arithmetic - algorithms and hardware designs , 1999 .

[22]  Michael Vogt,et al.  SMO Algorithms for Support Vector Machines without Bias Term , 2002 .

[23]  Lipo Wang.  Support vector machines : theory and applications , 2005 .

[24]  Luca Benini,et al.  Dynamic Reconfiguration in Sensor Networks with Regenerative Energy Sources , 2007, 2007 Design, Automation & Test in Europe Conference & Exhibition.

[25]  Michael I. Jordan,et al.  Nonparametric decentralized detection using kernel methods , 2005, IEEE Transactions on Signal Processing.

[26]  Dharma P. Agrawal,et al.  SVM-based intrusion detection system for wireless ad hoc networks , 2003, 2003 IEEE 58th Vehicular Technology Conference. VTC 2003-Fall (IEEE Cat. No.03CH37484).

[27]  Sven Leyffer,et al.  Numerical Experience with Lower Bounds for MIQP Branch-And-Bound , 1998, SIAM J. Optim..

[28]  D. Puccinelli,et al.  Wireless sensor networks: applications and challenges of ubiquitous sensing , 2005, IEEE Circuits and Systems Magazine.

[29]  Davide Anguita,et al.  Mixing floating- and fixed-point formats for neural network learning on neuroprocessors , 1996, Microprocess. Microprogramming.

[30]  David R. Musicant,et al.  Successive overrelaxation for support vector machines , 1999, IEEE Trans. Neural Networks.

[31]  Davide Anguita,et al.  Feed-Forward Support Vector Machine Without Multipliers , 2006, IEEE Transactions on Neural Networks.

[32]  Don R. Hush,et al.  QP Algorithms with Guaranteed Accuracy and Run Time for Support Vector Machines , 2006, J. Mach. Learn. Res..