Sparse Kernel Feature Analysis (SKFA). Numerical experi-

A fast algorithm, Accelerated Kernel Feature Analysis (AKFA), that discovers salient features evidenced in a sam-ple of n unclassified patterns, is presented. Like earlier kernel-based feature selection algorithms, AKFA implicitly embeds each pattern into a Hilbert space, H, induced by a Mercer kernel. An -dimensional linear subspace of H is iteratively constructed by maximizing a variance condi-tion for the nonlinearly transformed sample. This linear subspace can then be used to define more efficient data rep-resentations and pattern classifiers. AKFA requires O(n2) operations, as compared toO(n3) for Schölkof, Smola, an