Erasing automata recognize more than context-free languages
We consider the problem of estimating a function on for large values of by looking for some best approximation of by composite functions of the form . Our solution is based on model selection and leads to a very general approach to solve this problem with respect to many different types of functions and statistical frameworks. In particular, we handle the problems of approximating by additive functions, single and multiple index models, artificial neural networks, mixtures of Gaussian...
Pseudorandom binary sequences are required in stream ciphers and other applications of modern communication systems. In the first case it is essential that the sequences be unpredictable. The linear complexity of a sequence is the amount of it required to define the remainder. This work addresses the problem of the analysis and computation of the linear complexity of certain pseudorandom binary sequences. Finally we conclude some characteristics of the nonlinear function that produces the sequences...
The paper describes the background, the expected functionalities, and the architecture design goals of the European Digital Mathematics Library (Eu-DML), an infrastructure system aimed to integrate the mathematical contents available online throughout Europe, allowing for both extensive and specialized mathematics resource discovery. The three years long project to build the EuDML, partially funded by the European Commission, started in February 2010.