Relative Deviation Margin Bounds
Corinna Cortes 1 Mehryar Mohri 1 2 Ananda Theertha Suresh 1
Abstract they are used with positive definite kernels such as Gaus-
We present a series of new and more favorable sian kernels, for which the dimension of the feature space
margin-based learning guarantees that depend on is infinite. Similarly, margin-based learning bounds have
the empirical margin loss of a predictor. We give hel ...


雷达卡




京公网安备 11010802022788号







