Genetic Programming offers freedom in the definition of the cost function that is unparalleled among supervised learning algorithms. However, this freedom goes largely unexploited...
We obtain the strongest separation between quantum and classical query complexity known to date--specifically, we define a black-box problem that requires exponentially many querie...
We propose a probabilistic factorial sparse coder model for single channel source separation in the magnitude spectrogram domain. The mixture spectrogram is assumed to be the sum ...
Robert Peharz, Michael Stark, Franz Pernkopf, Yann...
This work deals with the power of linear algebra in the context of multilinear computation. By linear algebra we mean algebraic branching programs (ABPs) which are known to be com...
Zeev Dvir, Guillaume Malod, Sylvain Perifel, Amir ...
We study the extent to which properties of infinite-dimensional dynamical systems can be accurately detected by examining observations of such systems. Let H be a separable Hilbert...