Spectrum Estimation from a Few Entries

Authors: Ashish Khetan, Sewoong Oh

JMLR 2019 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical experiments suggest that we significantly improve upon a competing approach of using matrix completion methods, below the matrix completion threshold, above which matrix completion algorithms recover the underlying low-rank matrix exactly.
Researcher Affiliation Academia Department of Industrial and Enterprise Systems Engineering University of Illinois Urbana-Champaign Urbana, IL 61801, USA
Pseudocode Yes Algorithm 1 Schatten k-norm estimator
Open Source Code Yes A MATLAB implementation of the estimator (3), that includes as its sub-routines the computation of the weights of all k-cyclic pseudographs, is available for download at https://github.com/khetan2/Schatten_norm_estimation.
Open Datasets No The paper describes generating synthetic data for its numerical experiments, not the use of pre-existing public datasets. For example, in the description of Figure 4, it states: "M is a symmetric positive semi-definite matrix of size d = 500, and rank r = 100 (left panel) and r = 500 (right panel). Singular vectors U of M = UΣU , are generated by QR decomposition of N(0, Id d) and Σi,i is uniformly distributed over [1, 2]."
Dataset Splits No The paper describes theoretical analysis and numerical experiments primarily involving synthetically generated matrices and a sampling process (Erdos-Renyi sampling of matrix entries). It does not specify traditional training/test/validation dataset splits typically found in empirical machine learning studies for model training and evaluation.
Hardware Specification No The paper does not provide any specific hardware details (e.g., CPU/GPU models, memory specifications) used for running the experiments. It focuses on theoretical computational complexity (e.g., O(dα)).
Software Dependencies No The paper mentions "A MATLAB implementation of the estimator (3)" on page 8, but it does not specify a version number for MATLAB or any other critical software libraries or dependencies required to reproduce the experiments.
Experiment Setup Yes M is a symmetric positive semi-definite matrix of size d = 500, and rank r = 100 (left panel) and r = 500 (right panel). Singular vectors U of M = UΣU , are generated by QR decomposition of N(0, Id d) and Σi,i is uniformly distributed over [1, 2]. For a low rank matrix on the left, there is a clear critical value of p = 0.45, above which matrix completion is exact with high probability. We construct a symmetric matrix M of size d = 1000 and rank r = 200, σi Uni(0, 0.4) for 1 i r/2, and σi Uni(0.6, 1) for r/2 + 1 i r. We estimate br(PΩ(M); c1, c2) for Erd os-R enyi sampling Ω, and a choice of c2 = 0.5 and c1 = 0.6, which is motivated by the distribution of σi. We use Chebyshev polynomial of degree Cb = 2, and s = 1 for qs.