Nacionalni portal odprte znanosti

The proposed method uses homonymous and heteronymous examplepairs to train a linear preprocessor on a kernel-induced Hilbert space. The algorithm seeks to optimize the expected performance of elementary classifiers to be generated from single future training examples. The method is justified by PAC- ...

The talk presents a method of unsupervised learning from stationary, vector-valued processes. The method selects a subspace on the basis of an objective which can be used to bound the expected classification error for a family of tasks posessing a temporal continuity property. We prove bounds on the ...

We present an algorithm which uses example pairs of equal or unequal class labels to select a projection on a kernel-induced Hilbert space. A representation of .nite dimensional projections as bounded lin- ear functionals on a space of Hilbert-Schmidt operators is exploited to give bounds on the Rad ...

From concentration inequalities for the suprema of Gaussian or Rademacher processes an inequality is derived. It is applied to sharpen existing and to derive novel bounds on the empirical Rademacher complexities of unit balls in various norms appearing in the context of structured sparsity and multi ...