Boosting: From data to hardware using automatic implementation tool

We propose a method of automatic hardware implementation of a decision rule based on the Adaboost algorithm. We review the principles of the classification method and we evaluate its hardware implementation cost in term of FPGA's slice, using different weak classifiers based on the general concept of hyperrectangle. We show how to combine the weak classifiers in order to find an efficient trade-off between classification performances and hardware implementation cost. We present results obtained using examples coming from UCI databases.

[1]  Bernhard Schölkopf,et al.  Support Vector methods in learning and feature extraction , 1998 .

[2]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[3]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[4]  Miteran,et al.  4 - Classification géométrique par polytopes de contraintes. Performances et intégration , 1994 .

[5]  P. Gorria,et al.  Architectures for a real time classification processor , 1994, Proceedings of IEEE Custom Integrated Circuits Conference - CICC '94.

[6]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[7]  Demessie Girma,et al.  Artificial Neural Network Implementation on a Fine-Grained FPGA , 1994, FPL.

[8]  Fan Yang,et al.  Access control: adaptation and real-time implantation of a face recognition method , 2001 .

[9]  Y. Taright,et al.  FPGA implementation of a multilayer perceptron neural network using VHDL , 1998, ICSP '98. 1998 Fourth International Conference on Signal Processing (Cat. No.98TH8344).

[10]  Thomas G. Dietterich,et al.  An experimental comparison of the nearest-neighbor and nearest-hyperrectangle algorithms , 1995, Machine Learning.

[11]  Patrick Gorria,et al.  Classification géométrique par polytopes de contraintes , 1991 .

[12]  Gerhard Tröster,et al.  High-Level Area and Performance Estimation of Hardware Building Blocks on FPGAs , 2000, FPL.

[13]  Scott Hauck,et al.  The roles of FPGAs in reprogrammable systems , 1998, Proc. IEEE.

[14]  Jiri Matas,et al.  Support vector machines for face authentication , 2002, Image Vis. Comput..

[15]  S. Salzberg.  A nearest hyperrectangle learning method , 2004, Machine Learning.

[16]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[17]  Marti A. Hearst.  Trends & Controversies: Support Vector Machines , 1998, IEEE Intell. Syst..