Adrien Courtois
YOU?
Author Swipe
View article: Une étude en profondeur de l'apprentissage profond
Une étude en profondeur de l'apprentissage profond Open
Malgré l'immense succès des réseaux de neurones dans de nombreuses applications, la compréhension de leurs performances et de leurs limites progresse lentement. Bien que les théorèmes d'approximation universelle apportent la promesse de ma…
View article: SING: A Plug-and-Play DNN Learning Technique
SING: A Plug-and-Play DNN Learning Technique Open
We propose SING (StabIlized and Normalized Gradient), a plug-and-play technique that improves the stability and generalization of the Adam(W) optimizer. SING is straightforward to implement and has minimal computational overhead, requiring…
View article: Can neural networks extrapolate? Discussion of a theorem by Pedro Domingos
Can neural networks extrapolate? Discussion of a theorem by Pedro Domingos Open
View article: Video joint denoising and demosaicing with recurrent CNNs
Video joint denoising and demosaicing with recurrent CNNs Open
International audience
View article: Improving Pixel-Level Contrastive Learning by Leveraging Exogenous Depth Information
Improving Pixel-Level Contrastive Learning by Leveraging Exogenous Depth Information Open
Self-supervised representation learning based on Contrastive Learning (CL) has been the subject of much attention in recent years. This is due to the excellent results obtained on a variety of subsequent tasks (in particular classification…
View article: Can neural networks extrapolate? Discussion of a theorem by Pedro Domingos
Can neural networks extrapolate? Discussion of a theorem by Pedro Domingos Open
Neural networks trained on large datasets by minimizing a loss have become the state-of-the-art approach for resolving data science problems, particularly in computer vision, image processing and natural language processing. In spite of th…
View article: Investigating Neural Architectures by Synthetic Dataset Design
Investigating Neural Architectures by Synthetic Dataset Design Open
International audience
View article: Investigating Neural Architectures by Synthetic Dataset Design
Investigating Neural Architectures by Synthetic Dataset Design Open
Recent years have seen the emergence of many new neural network structures (architectures and layers). To solve a given task, a network requires a certain set of abilities reflected in its structure. The required abilities depend on each t…