Quantitative Propagation of Chaos for SGD in Wide Neural Networks

Authors: Valentin De Bortoli, Alain Durmus, Xavier Fontaine, Umut Simsekli

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We perform various experiments on real datasets to validate our theoretical results, assessing the existence of these two regimes on classification problems and illustrating our convergence results.
Researcher Affiliation Academia Valentin De Bortoli University of Oxford debortoli@stats.ox.ac.uk Alain Durmus Université Paris-Saclay alain.durmus@cmla.ens-cachan.fr Xavier Fontaine Université Paris-Saclay fontaine@cmla.ens-cachan.fr Umut Sim sekli LTCI, Télécom Paris, Institut Polytechnique de Paris umut.simsekli@telecom-paris.fr
Pseudocode No The paper does not contain any pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any links to open-source code or state that code will be made available.
Open Datasets Yes We focus on the classification task for two datasets: MNIST [41] and CIFAR-10 [42].
Dataset Splits No The paper mentions "training and test accuracies" but does not explicitly specify validation dataset splits.
Hardware Specification No The paper does not explicitly describe the hardware used for running its experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes If not specified, we set α = 0, M = 100, T = 100, γ = 1. ... We consider the following set of parameters α = 0, M = 100, T = 10000, γ = 0.1.