Abstract | ||
---|---|---|
This paper examines two-stage techniques for learning kernels based on a notion of alignment. It presents a number of novel theoretical, al- gorithmic, and empirical results for alignment- based techniques. Our results build on previous work by Cristianini et al. (2001), but we adopt a different definition of kernel alignment and significantly extend that work in several direc- tions: we give a novel and simple concentration bound for alignment between kernel matrices; show the existence of good predictors for ker- nels with high alignment, both for classification and for regression; give algorithms for learning a maximum alignment kernel by showing that the problem can be reduced to a simple QP; and re- port the results of extensive experiments with this alignment-based method in classification and re- gression tasks, which show an improvement both over the uniform combination of kernels and over other state-of-the-art learning kernel methods. |
Year | Venue | Keywords |
---|---|---|
2010 | ICML | kernel method |
Field | DocType | Citations |
Kernel smoother,Radial basis function kernel,Computer science,Tree kernel,Theoretical computer science,Polynomial kernel,Artificial intelligence,Kernel (linear algebra),Kernel embedding of distributions,Algorithm,Kernel method,Variable kernel density estimation,Machine learning | Conference | 80 |
PageRank | References | Authors |
2.18 | 15 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Corinna Cortes | 1 | 6574 | 1120.50 |
Mehryar Mohri | 2 | 4502 | 448.21 |
Afshin Rostamizadeh | 3 | 911 | 44.15 |