On Single-Index Models beyond Gaussian Data

Authors: Aaron Zweig, Loucas PILLAUD-VIVIEN, Joan Bruna

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In order to validate our theory, and inspect the degree to which our bounds may be pessimistic, we consider empirical evaluation of the training process in our two primary settings. Specifically, we consider random initialization on the half-sphere (with the sign chosen to induce positive correlation as in Arous et al. [2021]), and investigate how often strong recovery occurs relative to the information exponent of the link function.
Researcher Affiliation Academia Joan Bruna New York University Loucas Pillaud-Vivien New York University Flatiron Institute Aaron Zweig New York University
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any concrete access information (e.g., repository link or explicit statement of code release) for the source code of the methodology described.
Open Datasets No The paper does not provide concrete access information (specific link, DOI, repository name, formal citation, or reference to established benchmark datasets) for a publicly available or open dataset. It discusses data distributions conceptually but does not specify a named dataset used in experiments with access details.
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) for training, validation, or test sets.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment.
Experiment Setup Yes Specifically, we consider random initialization on the half-sphere (with the sign chosen to induce positive correlation as in Arous et al. [2021]).