A Differential Theory of Learning for Efficient Statistical Pattern Recognition
Final rept. 30 Sep 1989-29 Sep 1993
CARNEGIE-MELLON UNIV PITTSBURGH PA
Pagination or Media Count:
Probabilistic learning strategies currently use are inefficient, requiring high classifier complexity and large training samples. In this report, we introduce and analyze an asymptotically efficient differential learning strategy. It guarantees the best generalization allowed by the chosen classifier paradigm. Differential learning also requires the classifier with minimal complexity. The theory is demonstrated in several real-world machine learningpattern recognition tasks. Learning, Pattern recognition, Classification, Neural networks.
- Statistics and Probability