Rational Kernels: Theory and Algorithms

By Corinna Cortes, Patrick Haffner, Mehryar Mohri. 2004

206 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Generative and Latent Mean Map Kernels (2010, background): “Rational kernels [3] are another class of kernels that can handle variable-length sequences.”
Learning Deep Architectures for AI (2007, background): “The limitations of a fixed generic kernel such as the Gaussian kernel have motivated a lot of research in designing kernels based on prior knowledge about the task [38, 56, 89, 167].”
Neural Networks for Sequential Data: a Pre-training Approach based on Hidden Markov Models (2015, methodology): “, [5,6]); (iii) model-based approaches, such as using Hidden Markov Models (e.”
Nonextensive Information Theoretic Kernels on Measures (2009, background): “, kernels operating on the space of strings) have been proposed in the literature (Haussler, 1999; Lodhi et al., 2002; Leslie et al., 2002; Vishwanathan and Smola, 2003; Shawe-Taylor and Cristianini, 2004; Cortes et al., 2004).”
Support Vector Machines and Kernels for Computational Biology (2008, background): “Kernels (and hence the similarity) on structured data can also be understood as how much one object has to be transformed before it is identical to the other, which leads to the idea of transducers [55].”
Identifying gene-disease associations using centrality on a literature mined gene-interaction network (2008, background): “Cortes et al. (2004) proved that the edit kernel is not always positive definite.”
A general regression technique for learning transductions (2005, methodology): “These kernels are instances ofrational kernelsand have been used successfully in a variety of difficult prediction tasks in text and speech processing (Cortes et al., 2004).”
Cancer Evolution: Mathematical Models and Computational Inference (2014, unknown): “Based on finite-state transducers (Cortes et al. 2004; Schwarz et al. 2010), MEDICC computes the minimum number of amplification and deletion events to transform one genomic profile into another.”