Fisher Kernel

The Fisher kernel is defined as


K(X_i, X_j) = U_{X_i}^T I^{-1} U_{X_j}

with I the Fisher information matrix.

Famous quotes containing the words fisher and/or kernel:

    ... most bereaved souls crave nourishment more tangible than prayers: they want a steak. What is more, they need a steak. Preferably they need it rare, grilled, heavily salted, for that way it is most easily digested, and most quickly turned into the glandular whip their tired adrenals cry for.
    —M.F.K. Fisher (1908–1992)

    All true histories contain instruction; though, in some, the treasure may be hard to find, and when found, so trivial in quantity that the dry, shrivelled kernel scarcely compensates for the trouble of cracking the nut.
    Anne Brontë (1820–1849)