Similarity metric learning for a variable-kernel classifier
Neural Computation
Classification Using a Hierarchical Bayesian Approach
ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 4 - Volume 4
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Classification by probabilistic clustering
ICASSP '01 Proceedings of the Acoustics, Speech, and Signal Processing, 200. on IEEE International Conference - Volume 02
Learning similarity measures: a formal view based on a generalized CBR model
ICCBR'05 Proceedings of the 6th international conference on Case-Based Reasoning Research and Development
Bounding the probability of error for high precision optical character recognition
The Journal of Machine Learning Research
The Journal of Machine Learning Research
Hi-index | 0.00 |
Handwriting recognition and OCR systems needto cope with a wide variety of writing styles and fonts,many of them possibly not previously encountered duringtraining. This paper describes a notion of Bayesian statisticalsimilarity and demonstrates how it can be appliedto rapid adaptation to new styles. The ability to generalizeacross different problem instances is illustrated in theGaussian case, and the use of statistical similarity Gaussiancase is shown to be related to adaptive metric classificationmethods. The relationship to prior approaches tomultitask learning, as well as variable or adaptive metricclassification, and hierarchical Bayesian methods, are discussed.Experimental results on character recognition fromthe NIST3 database are presented.