Asynchronous Stochastic Proximal Optimization Algorithms with Variance Reduction

Authors: Qi Meng, Wei Chen, Jingcheng Yu, Taifeng Wang, Zhi-Ming Ma, Tie-Yan Liu

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We have conducted experiments on a regularized logistic regression task. The results verified our theoretical findings and demonstrated the practical efficiency of the asynchronous stochastic proximal algorithms with variance reduction.
Researcher Affiliation Collaboration Qi Meng,1 Wei Chen,2 Jingcheng Yu,3 Taifeng Wang,2 Zhi-Ming Ma,4 Tie-Yan Liu2 1 School of Mathematical Sciences, Peking University, qimeng13@pku.edu.cn 2Microsoft Research, {wche, taifengw, tie-yan.liu}@microsoft.com 3Carnegie Mellon University, jingchey@cs.cmu.edu 4Academy of Mathematics and Systems Science, Chinese Academy of Sciences, mazm@amt.ac.cn
Pseudocode Yes Algorithm 1 Async-Prox SVRG and Async-Prox SVRCD
Open Source Code No The paper does not provide an explicit statement of source code release or a link to a code repository for the methodology described.
Open Datasets Yes We conducted binary classifications on three benchmark datasets: rcv1, real-sim, news20 (Reddi et al. 2015)
Dataset Splits No The paper mentions dataset sizes but does not provide specific training, validation, or test split percentages or counts needed to reproduce the data partitioning.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment.
Experiment Setup Yes For Async-Prox SVRG, we set step size η = 0.04, the minibatch size B = 200, and the inner loop size K = 2n, where n is the data size. For Async-Prox SVRCD, we set step size η = 0.04, the number of block partitions m = d 100, the minibatch size B = 200, and a larger inner loop size K = 2nm.