A Tight Bound of Hard Thresholding

Authors: Jie Shen, Ping Li

JMLR 2017 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we present a comprehensive empirical study for the proposed HT-SVRG algorithm on two tasks: sparse recovery (compressed sensing) and image classification. The experiments on sparse recovery is dedicated to verifying the theoretical results we presented, and we visualize the classification models learned by HT-SVRG to demonstrate the practical efficacy.
Researcher Affiliation Collaboration Jie Shen EMAIL Department of Computer Science Rutgers University Piscataway, NJ 08854, USA Ping Li EMAIL Baidu Research Bellevue, WA 98004, USA
Pseudocode Yes Algorithm 1 Hard Thresholded Stochastic Variance Reduced Gradient Method (HT-SVRG)
Open Source Code No No explicit statement or link to source code for the methodology described in this paper is provided. The paper only mentions a concurrent work by Li et al. (2016) that made their work public.
Open Datasets Yes Here, we study the performance on a realistic image dataset MNIST1, consisting of 60 thousands training samples and 10 thousands samples for testing. There is one digit on each image of size 28-by-28, hence totally 10 classes. Some of the images are shown in Figure 6. The update frequency m is fixed as m = 3n. We compute the heuristic step size η as in the previous section, i.e., η = 2/svds(AA ) 10 3. Since for the real-world dataset, the true sparsity is actually unknown, we tune the sparsity parameter k and study the performance of the algorithm. 1. http://yann.lecun.com/exdb/mnist/
Dataset Splits Yes Here, we study the performance on a realistic image dataset MNIST1, consisting of 60 thousands training samples and 10 thousands samples for testing.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments are provided in the paper.
Software Dependencies No No specific software dependencies with version numbers (e.g., library or solver names with version numbers) are mentioned in the paper.
Experiment Setup Yes If not specified, we use m = 3n, k = 9K, and S = 10000 for HT-SVRG. We also use the heuristic step size η = 2/svds(AA ) for HT-SVRG and PGD, where svds(AA ) returns the largest singular value of the matrix AA . For the MNIST dataset: "The update frequency m is fixed as m = 3n. We compute the heuristic step size η as in the previous section, i.e., η = 2/svds(AA ) 10 3."