no code implementations • 15 Mar 2022 • Simon Carbonnelle
Hence, this thesis attempts to identify implicit clustering abilities, mechanisms and hyperparameters in deep learning systems and evaluate their relevance for explaining the generalization abilities of these systems.
1 code implementation • ICML Workshop Deep_Phenomen 2019 • Simon Carbonnelle, Christophe De Vleeschouwer
Our work presents empirical evidence that layer rotation, i. e. the evolution across training of the cosine distance between each layer's weight vector and its initialization, constitutes an impressively consistent indicator of generalization performance.
no code implementations • 27 Sep 2018 • Simon Carbonnelle, Christophe De Vleeschouwer
How optimization influences the generalization ability of a DNN is still an active area of research.
2 code implementations • 5 Jun 2018 • Simon Carbonnelle, Christophe De Vleeschouwer
Our work presents extensive empirical evidence that layer rotation, i. e. the evolution across training of the cosine distance between each layer's weight vector and its initialization, constitutes an impressively consistent indicator of generalization performance.
no code implementations • ICLR 2018 • Simon Carbonnelle, Christophe De Vleeschouwer
Neural networks trained through stochastic gradient descent (SGD) have been around for more than 30 years, but they still escape our understanding.