Theses and dissertations

  • Accreditation to supervise research (HDR) (download)

I defended an accreditation to supervise research in october 2018, at the University Bourgogne Franche-Comté, entitled On the use of neural networks to solve problems. From multilayer perceptron to deep learning and reservoir computing, which has the following abstract:

Since the introduction of the perceptron at the end of the 1950s, neural networks have passed through several boom-and-bust episodes, becoming today the state-of-the-art approach for solving many classification and prediction problems. The backpropagation algorithm has thus given rise to the multilayer networks era. More recently, there has been a revival with the emergence of various deep neural networks also known as deep learning architectures. The research works described in this manuscript reflect on the one hand this network architecture evolution and on the other hand highlight the suitability of neural networks for successfully solving tasks in different areas. In a first part we show that a multilayer perceptron neural network can simulate lung motion being able to predict the trajectory of several feature points, that it can be a substitute for a computational fluid dynamics software for active airflow control, and finally allows to build also a true chaotic neural network. In the second part, convolutional neural networks (a main flavor of deep learning) are used to detect whether an image embeds a hidden message or not, the obtained steganalysis approach was at the time of its publication the state-of-the-art one. Then we propose a fully convolutional neural network for image denoising, this latter, which is an encoder-decoder, can deal with different kinds of noise. Finally, considering, the Reservoir Computing paradigm studied by colleagues of our research institute, we study the application of Echo State Networks, a recurrent architecture, on a handwritten digits recognition problem (namely the well-known MNIST benchmark problem).

I defended a thesis in computer science in december 2001, at the (Louis Pasterur) University of Strabourg, entitled Parallelizing heuristics for combinatorial optimization: a study in the context of medical image registration, under the supervision of Professsors Guy-René Perrinf and Fabrice Heitz. In this work I have mainly studied the application of the data-parallel model to implement global optimization algorithms. Two algorithms were considered: differential evolution and form of Langevin's stochastic differential equation. Differential evolution has been applied to rigid image matching, whereas the second algorithm was focused on three dimensional deformable image matching. In both cases the considered images where MRI with a size of  1283 or 2563 voxels.

During my internship I studied the classical wavelet-based image compression scheme: computation of wavelets coefficients followed by their quantization, and finally entropic coding; as well as two algorithms that thanks to quantization strategies through successive approximations  allow to perform progressive transmission. A data-parallel version of the classical pyramid algorithm for computing the discrete wavelet transform (DWT) has been in particular designed and evaluated.