Least Squares Support Vector Machine Classifiers
Neural Processing Letters
Sparse Greedy Matrix Approximation for Machine Learning
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
The Effect of the Input Density Distribution on Kernel-based Classifiers
ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Efficient svm training using low-rank kernel representations
The Journal of Machine Learning Research
Kernel independent component analysis
The Journal of Machine Learning Research
Kernel Methods for Pattern Analysis
Kernel Methods for Pattern Analysis
Learning the Kernel Matrix with Semidefinite Programming
The Journal of Machine Learning Research
Learning low-rank kernel matrices
ICML '06 Proceedings of the 23rd international conference on Machine learning
Incremental approximate matrix factorization for speeding up support vector machines
Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
Computational Statistics & Data Analysis
Building Support Vector Machines with Reduced Classifier Complexity
The Journal of Machine Learning Research
Rank minimization via online learning
Proceedings of the 25th international conference on Machine learning
Improved Nyström low-rank approximation and error analysis
Proceedings of the 25th international conference on Machine learning
Sparse kernel SVMs via cutting-plane training
Machine Learning
Feature Selection for Value Function Approximation Using Bayesian Model Selection
ECML PKDD '09 Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: Part I
On speeding up computation in information theoretic learning
IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
On-line independent support vector machines
Pattern Recognition
Sparse approximation through boosting for learning large scale kernel machines
IEEE Transactions on Neural Networks
Clustered Nyström method for large scale manifold learning and dimension reduction
IEEE Transactions on Neural Networks
Using an iterative linear solver in an interior-point method for generating support vector machines
Computational Optimization and Applications
Random Fourier approximations for skewed multiplicative histogram kernels
Proceedings of the 32nd DAGM conference on Pattern recognition
Mixing linear SVMs for nonlinear classification
IEEE Transactions on Neural Networks
Regression on Fixed-Rank Positive Semidefinite Matrices: A Riemannian Approach
The Journal of Machine Learning Research
Multi-subspace representation and discovery
ECML PKDD'11 Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part II
Sequential learning with LS-SVM for large-scale data sets
ICANN'06 Proceedings of the 16th international conference on Artificial Neural Networks - Volume Part II
Expert Systems with Applications: An International Journal
Online independent reduced least squares support vector regression
Information Sciences: an International Journal
ICAISC'12 Proceedings of the 11th international conference on Artificial Intelligence and Soft Computing - Volume Part II
Cluster indicator decomposition for efficient matrix factorization
IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Two
Learning low-rank Mercer kernels with fast-decaying spectrum
Neurocomputing
Sampling methods for the Nyström method
The Journal of Machine Learning Research
Learning with infinitely many features
Machine Learning
Reduced heteroscedasticity linear regression for Nyström approximation
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
Training sparse SVM on the core sets of fitting-planes
Neurocomputing
| Hi-index | 0.00 |
Low-rank matrix decompositions are essential tools in the application of kernel methods to large-scale learning problems. These decompositions have generally been treated as black boxes---the decomposition of the kernel matrix that they deliver is independent of the specific learning task at hand---and this is a potentially significant source of inefficiency. In this paper, we present an algorithm that can exploit side information (e.g., classification labels, regression responses) in the computation of low-rank decompositions for kernel matrices. Our algorithm has the same favorable scaling as state-of-the-art methods such as incomplete Cholesky decomposition---it is linear in the number of data points and quadratic in the rank of the approximation. We present simulation results that show that our algorithm yields decompositions of significantly smaller rank than those found by incomplete Cholesky decomposition.