Geometry and Stability of Supervised Learning Problems

Authors: Facundo Mémoli, Brantley Vose, Robert C. Williamson

JMLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical The utility of the Risk distance lies not in explicit algorithms and associated computations, but in the theoretical landscape that it facilitates. The paper is primarily focused on introducing a new mathematical notion of distance, exploring its geometric properties, and proving stability results, propositions, and theorems.
Researcher Affiliation Academia All listed authors are affiliated with universities: Rutgers University, The Ohio State University, and University of Tubingen. The email domains (.gmail.com for one author is a personal email, but the institutional affiliation is academic; .osu.edu and .uni-tuebingen.de are academic domains) also support academic affiliations.
Pseudocode No The paper does not contain any sections explicitly labeled 'Pseudocode' or 'Algorithm', nor does it present structured steps for a method in a code-like format. Its content is primarily theoretical definitions, propositions, and proofs.
Open Source Code No The paper does not provide any statements about releasing code, nor does it include links to source code repositories or mention code in supplementary materials.
Open Datasets No The paper introduces a theoretical framework and uses abstract examples and toy models (e.g., Example 18, 19, 40, 53, 90, 92) to illustrate concepts, but does not use or provide access to any publicly available datasets for experimental validation.
Dataset Splits No As the paper does not conduct empirical experiments with datasets, there is no mention of dataset splits.
Hardware Specification No The paper focuses on theoretical contributions and does not describe any specific hardware used for running simulations or analyses.
Software Dependencies No The paper discusses theoretical concepts and mathematical proofs, and does not specify any software dependencies with version numbers for implementation or experimentation.
Experiment Setup No Given the theoretical nature of the paper, there are no experimental setups, hyperparameters, or training configurations described.