Nearly Isometric Embedding by Relaxation

Authors: James McQueen, Marina Meila, Dominique Joncas

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experiments confirm the superiority of our algorithm in obtaining low distortion embeddings. 4 Experimental evaluation
Researcher Affiliation Collaboration James Mc Queen Department of Statistics University of Washington Seattle, WA 98195 jmcq@u.washington.edu Marina Meil a Department of Statistics University of Washington Seattle, WA 98195 mmp@stat.washington.edu Dominique Perrault-Joncas Google Seattle, WA 98103 dcpjoncas@gmail.com
Pseudocode Yes Algorithm 1: Outline of the Riemannian Relaxation Algorithm. Algorithm 2: RIEMANNIANRELAXATION (RR) Algorithm 3: PRINCIPALCURVES-RIEMANNIANRELAXATION (PCS-RR)
Open Source Code No The paper does not provide an explicit statement about the release of its own source code or a link to a repository for the methodology described.
Open Datasets Yes The data consists of spectra of galaxies from the Sloan Digital Sky Survey7 [1]. 7 www.sdss.org
Dataset Splits No The paper describes total sample sizes used in experiments (e.g., n = 10000, n = 3000, n = 2000 subsample) but does not provide explicit training, validation, or test set split percentages or counts.
Hardware Specification No The paper does not provide specific details regarding the hardware used for running the experiments, such as GPU or CPU models, or cloud computing instance types.
Software Dependencies No The paper mentions the use of 'drtoolbox' for computing embeddings of other algorithms, but it does not specify version numbers for this or any other software dependencies.
Experiment Setup Yes Algorithm 2 mentions 'heavy ball parameter α [0, 1)' and finding 'step size η by line search'. Section 4 states 'Convergence of RR was achieved after 400 iterations' and 'Convergence of PCS-RR was achieved after 1000 iterations'.