Averaging Weak Classifiers
We present a learning algorithm for two-class pattern recognition. It is based on combining a large number of weak classifiers. The weak classifiers are produced independently with diversity. And they are combined through a weighted average, weighted exponentially with respect to their apparent errors on the training data. Experimental results are also given.
KeywordsFeature Vector Test Error Rectangular Region Weak Classifier Test Error Rate
Unable to display preview. Download preview PDF.
- 2.Chen, D. and Cheng, X. (2000, a). A simple implementation of the stochastic discrimination for pattern recognition. In Proc. Joint IAPR International Workshops SSPR 2000 and SPR 2000 (Francesc J. Ferri, José M. Iñesta, et al., eds.), Alicante, Spain, Springer LNCS-1876, pp. 882–887.Google Scholar
- 3.Chen, D. and Cheng, X. (2000, b). Majority vote based on weak classifiers. 2000 5th International Conference on Signal Processing Proceedings, Beijing, China, vol. 3, pp. 1560–1562.Google Scholar
- 4.Freund, Y. and Schapire, R. E. (1996). Experiments with a new boosting algorithm. Proc. 13th Int’l Conf. Machine Learning, pp. 148–156.Google Scholar
- 5.Freund, Y., Mansour, Y., Schapire, R. E. (2001). Why averaging classifiers can protect against overfitting. In Proceedings of the Eighth International Workshop on Artificial Intelligence and Statistics.Google Scholar
- 11.Kleinberg, E. M. (2000, b). A mathematically rigorous foundation for supervised learning. Proc. First Int’l Workshop on Multiple Classifier Systems, Cagliari, Italy, pp. 67–76.Google Scholar
- 12.MacKay, D. J. C. (1991). Bayesian Methods for Adaptive Models, Ph.D. thesis, California Institute of Technology.Google Scholar
- 13.Smith, J., Everhart, J., Dickson, W., Knowler, W., & Johannes, R. (1988). Using the ADAP learning algorithm to forecast the onset of diabetes mellitus. In Proceedings of the Symposium on Computer Applications and Medical Care, IEEE Computer Society Press, pp. 261–265.Google Scholar