We consider generalized linear models where an unknown $n$-dimensional signal vector is observed through the successive application of a random matrix and a non-linear (possibly probabilistic) componentwise function. We consider the models in the high-dimensional limit, where the observation consists of $m$ points, and $m/n {\to} {\alpha}$ where ${...
Finding a solution to a Constraint Satisfaction Problem (CSP) is known to be an NP-complete task. This has motivated the multitude of works that have been devoted to developing techniques that simplify CSP instances before or during their resolution. The present work proposes rigidly enforced schemes for simplifying binary CSPs that allow the narro...
We examine a network of learners which address the same classification task but must learn from different data sets. The learners can share a limited portion of their data sets so as to preserve the network load. We introduce DELCO (standing for Decentralized Ensemble Learning with COpulas), a new approach in which the shared data and the trained m...
One of the most burning issue in deep learning community is the well known adversarial example phenomenon where two very close samples are classified differently. In this paper, I offer to see in this phenomenon the opportunity to help the training with a new kind of regularisation and/or semi supervised training. The main idea is to use both class...
Published in Materials (Basel, Switzerland)
A new treatment method of combined crumb rubber and steel slag modifier for asphalt binders was proposed in this work. The viscosity, rheological properties, and thermogravimetric analysis of modified asphalt mortar were then investigated. The modified asphalt composite was prepared in laboratory by two steps. Rubber powder was firstly added into h...
This paper is a reflexion on the computability of natural language semantics. It does not contain a new model or new results in the formal semantics of natural language: it is rather a computational analysis of the logical models and algorithms currently used in natural language semantics, defined as the mapping of a statement to logical formulas —...
Training a Deep Neural Network (DNN) from scratch requires a large amount of labeled data. For a classification task where only small amount of training data is available, a common solution is to perform fine-tuning on a DNN which is pre-trained with related source data. This consecutive training process is time consuming and does not consider expl...
Deep learning is such a breakthrough that deep networks may quickly be allowed to take critical decisions (diagnosis, really autonomous driving).Yet, today, there is no process, both safe and business compliant, to evaluate a deep learning system.Worse, I present experiments showing that such process is even father than one could have thought.
We present an efficient deep learning approach for the challenging task of tumor segmentation in multisequence MR images. In recent years, Convolutional Neural Networks (CNN) have achieved state-of-the-art performances in a large variety of recognition tasks in medical imaging. Because of the considerable computational cost of CNNs, large volumes s...
