Deriving kernels from MLP probability estimators for large categorization problems

I. Titov, J. Henderson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

In multi-class categorization problems with a very large or unbounded number of classes, it is often not computationally feasible to train and/or test a kernel-based classifier. One solution is to use a fast computation to pre-select a subset of the classes for reranking with a kernel method, but even then tractability can be a problem. We investigate using trained multilayer perceptron probability estimators to derive appropriate kernels for such problems. We propose a kernel derivation method which is specifically designed for reranking problems, and a more efficient variant of this method which is specifically designed for neural networks with large numbers of output units. When applied to a neural network model of natural language parsing, these new methods achieve state-of-the-art performance which improves over the original model.
Original languageEnglish
Title of host publicationProceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005.
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages937-942 vol. 2
Volume2
ISBN (Print)0-7803-9048-2
DOIs
Publication statusPublished - 1 Jul 2005

Fingerprint

Dive into the research topics of 'Deriving kernels from MLP probability estimators for large categorization problems'. Together they form a unique fingerprint.

Cite this