Improved Particle Approximation Error for Mean Field Neural Networks

Authors: Atsushi Nitanda

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this work, we improve the dependence on logarithmic Sobolev inequality (LSI) constants in their particle approximation errors which can exponentially deteriorate with the regularization coefficient. Specifically, we establish an LSI-constant-free particle approximation error concerning the objective gap by leveraging the problem structure in risk minimization. As the application, we demonstrate improved convergence of MFLD, sampling guarantee for the mean-field stationary distribution, and uniform-in-time Wasserstein propagation of chaos in terms of particle complexity.
Researcher Affiliation Collaboration Atsushi Nitanda CFAR and IHPC, Agency for Science, Technology and Research (A STAR), Singapore College of Computing and Data Science, Nanyang Technological University, Singapore atsushi_nitanda@cfar.a-star.edu.sg
Pseudocode No The paper describes mathematical derivations and proofs but does not include any pseudocode or algorithm blocks.
Open Source Code No The paper is theoretical and does not mention releasing any source code or provide links to a code repository. The NeurIPS checklist confirms that the paper does not include experiments, implying no code release for such.
Open Datasets No This paper is theoretical and does not involve experiments or the use of datasets.
Dataset Splits No This paper is theoretical and does not involve experiments or specify dataset splits for training, validation, or testing.
Hardware Specification No This paper is theoretical and does not include experiments that would require hardware specifications. The NeurIPS checklist (Question 8, Answer NA) confirms this.
Software Dependencies No This paper is theoretical and does not include experiments that would require specific software dependencies with version numbers. The NeurIPS checklist (Question 6, Answer NA) confirms this.
Experiment Setup No This paper is theoretical and does not include experiments or describe an experimental setup. The NeurIPS checklist (Question 6, Answer NA) confirms this.