We consider a general class of regularization methods which learn a vector of parameters on the basis of linear measurements. It is well known that if the regularizer is a nondecr...
Andreas Argyriou, Charles A. Micchelli, Massimilia...
Fixpoint expressions built from functional signatures interpreted over arbitrary complete lattices are considered. A generic notionof automatonis defined and shown, by means of a ...
We give several versions of local and global inverse mapping theorem for tame non necessarily smooth, mappings. Here tame mapping means a mapping which is subanalytic or, more gene...
Toshizumi Fukui, Krzysztof Kurdyka, Laurentiu Paun...
We establish several approximate max-integral-flow / minmulticut theorems. While in general this ratio can be very large, we prove strong approximation ratios in the case where th...
In this paper we present a method for reasoning about privacy using the concepts of exchangeability and deFinetti's theorem. We illustrate the usefulness of this technique by...