Related Experiment Videos
A Sparse Representation for Function Approximation
1Massachusetts Institute of Technology, Artificial Intelligence Lab, Cambridge MA, US, 545 Technology Square, 02139. tp@ai.mit.edu
Neural Computation
|August 11, 1998
Abstract:
We derive a new general representation for a function as a linear combination of local correlation kernels at optimal sparse locations (and scales) and characterize its relation to principal component analysis, regularization, sparsity principles, and support vector machines.