A learning algorithm based on primary school teaching wisdom

A learning algorithm based on primary school teaching and learning is presented. The methodology is to continuously evaluate the performance of the network and to train it on the examples for which they repeatedly fail, until all the examples are correctly classified. Empirical analysis on UCI data show that the algorithm produces good training data and improves the generalization ability of the network on unseen data. The algorithm has interesting applications in data mining, model evaluations and rare objects discovery.

[1]  Slava M. Katz,et al.  Estimation of probabilities from sparse data for the language model component of a speech recognizer , 1987, IEEE Trans. Acoust. Speech Signal Process..

[2]  Masashi Sugiyama,et al.  Training Data Selection for Optimal Generalization in Trigonometric Polynomial Networks , 1999, NIPS.

[3]  J. Berger,et al.  Training samples in objective Bayesian model selection , 2004, math/0406460.

[4]  S. C. Odewahn,et al.  Automated Galaxy Morphology: A Fourier Approach , 2002 .

[5]  A. Dawid.  Conditional Independence in Statistical Theory , 1979 .

[6]  Dennis L. Wilson,et al.  Asymptotic Properties of Nearest Neighbor Rules Using Edited Data , 1972, IEEE Trans. Syst. Man Cybern..

[7]  Jayant Murthy,et al.  A three‐dimensional automated classification scheme for the TAUVEX data pipeline , 2007 .

[8]  M. H. Choueiki,et al.  Training data development with the D-optimality criterion , 1999, IEEE Trans. Neural Networks.

[9]  Kenji Nakayama,et al.  A training method with small computation for classification , 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.

[10]  Mark A. Fanty,et al.  Improved probability estimation with neural network models , 1996, Proceeding of Fourth International Conference on Spoken Language Processing. ICSLP '96.

[11]  Nir Friedman,et al.  Bayesian Network Classifiers , 1997, Machine Learning.

[12]  Benjamin Quost.  One-against-all Classifier Combination in the Framework of Belief Functions , 2006 .

[13]  Christoph F. Eick,et al.  Supervised clustering - algorithms and benefits , 2004, 16th IEEE International Conference on Tools with Artificial Intelligence.

[14]  Ajit Kembhavi,et al.  A difference boosting neural network for automated star-galaxy classification , 2002 .

[15]  Ninan Sajeeth Philip,et al.  Boosting the differences: A fast Bayesian classifier neural network , 2000, Intell. Data Anal..

[16]  Leon N. Cooper,et al.  Training Data Selection for Support Vector Machines , 2005, ICNC.

[17]  Francisco Herrera,et al.  On the combination of evolutionary algorithms and stratified strategies for training set selection in data mining , 2006, Appl. Soft Comput..

[18]  Charles Elkan,et al.  Boosting and Naive Bayesian learning , 1997 .

[19]  Lou Boves,et al.  In search of optimal data selection for training of automatic speech recognition systems , 2003, 2003 IEEE Workshop on Automatic Speech Recognition and Understanding (IEEE Cat. No.03EX721).

[20]  Ninan Sajeeth Philip,et al.  Advances in Automated Algorithms For Morphological Classification of Galaxies Based on Shape Features , 2004 .

[21]  M. Wann,et al.  The influence of training sets on generalization in feed-forward neural networks , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[22]  Huan Liu,et al.  Neural-network feature selector , 1997, IEEE Trans. Neural Networks.

[23]  Carlos Eduardo Pedreira,et al.  Learning vector quantization with training data selection , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[24]  Masashi Sugiyama,et al.  Training Data Selection for Optimal Generalization with Noise Variance Reduction in , 1998 .