Fast-and-Light Stochastic ADMM

Authors: Shuai Zheng, James T. Kwok

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

Reproducibility Variable Result LLM Response
Research Type Experimental 4 Experiments
Researcher Affiliation Academia Shuai Zheng James T. Kwok Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong {szhengac, jamesk}@cse.ust.hk
Pseudocode Yes Algorithm 1 SVRG-ADMM for strongly convex problems.
Open Source Code No No explicit statement regarding the release of source code for the described methodology or a direct link to a code repository is provided.
Open Datasets Yes Downloaded from http://www.csie.ntu.edu.tw/ cjlin/ libsvmtools/datasets/, http://osmot.cs.cornell.edu/kddcup/datasets. html, and http://largescale.ml.tu-berlin.de/instructions/.
Dataset Splits No The paper provides training and testing set sizes (e.g., 'We use 1,281,167 images for training, and 50, 000 images for testing' and Table 2), but does not explicitly detail a separate validation split or the methodology for creating these splits (e.g., random seed, stratified splitting).
Hardware Specification Yes Experiments are performed on a PC with Intel i7-3770 3.4GHz CPU and 32GB RAM
Software Dependencies No The paper mentions 'Matlab' as the environment used for comparison, but does not provide specific version numbers for Matlab or any other key software libraries/dependencies.
Experiment Setup Yes We use a mini-batch size of b = 100 on protein and covertype; and b = 500 on mnist8m and dna. The proposed SVRG-ADMM uses the linearized update in (12) and m = 2n/b. For SVRG-ADMM, since the learning rate in (12) is effectively /γ, we set γ = 1 and only tune . All parameters are tuned as in [Zhong and Kwok, 2014]. We set λ1 = 10 5, λ2 = 10 4, and use a mini-batch size b = 500. We set n = 100, 000, d = 500, λ = 0.1/pn, and a minibatch size b = 100.