Learning Bounds for Greedy Approximation with Explicit Feature Maps from Multiple Kernels

Open in new window