Replicability in Learning: Geometric Partitions and KKM-Sperner Lemma

Authors: Jason Vander Woude, Peter Dixon, A. Pavan, Jamie Radcliffe, N. V. Vinodchandran

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

Reproducibility Variable Result LLM Response
Research Type Theoretical This paper studies replicability in machine learning tasks from a geometric viewpoint. Recent works have revealed the role of geometric partitions and Sperner s lemma (and its variations) in designing replicable learning algorithms and in establishing impossibility results. ... Our first contribution is a comprehensive understanding of the optimality of secluded partition constructions. Our second contribution is the discovery of a new neighborhood variant of the Sperner/KKM lemma.
Researcher Affiliation Collaboration Jason Vander Woude Sandia National Laboratories jasonvwoude@gmail.com Peter Dixon University of Toronto, Mississauga tooplark@gmail.com A. Pavan Iowa State University pavan@cs.iastate.edu Jamie Radcliffe University of Nebraska-Lincoln jamie.radcliffe@unl.edu N. V. Vinodchandran University of Nebraska-Lincoln vinod@unl.edu
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No Our paper does not include experiments requiring code. (From NeurIPS checklist answer)
Open Datasets No Our paper does not include experiments. (From NeurIPS checklist answer)
Dataset Splits No Our paper does not include experiments. (From NeurIPS checklist answer)
Hardware Specification No Our paper does not include experiments. (From NeurIPS checklist answer)
Software Dependencies No Our paper does not include experiments. (From NeurIPS checklist answer)
Experiment Setup No Our paper does not include experiments. (From NeurIPS checklist answer)