NESTT: A Nonconvex Primal-Dual Splitting Method for Distributed and Stochastic Optimization

Authors: Davood Hajinezhad, Mingyi Hong, Tuo Zhao, Zhaoran Wang

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

Reproducibility Variable Result LLM Response
Research Type Experimental 5 Numerical Results In this section we evaluate the performance of NESTT. Consider the high dimensional regression problem with noisy observation [16], where M observations are generated by y = Xν + ϵ. ... In Fig. 1 we compare different algorithms in terms of the gap 1/βf(zr) 2. ... In Table 2 we further compare different algorithms when changing the number of component functions (i.e., the number of minibatches N) while the rest of the setup is as above.
Researcher Affiliation Academia Department of Industrial & Manufacturing Systems Engineering and Department of Electrical & Computer Engineering, Iowa State University, Ames, IA, {dhaji,mingyi}@iastate.edu School of Industrial and Systems Engineering, Georgia Institute of Technology tourzhao@gatech.edu Department of Operations Research, Princeton University,zhaoran@princeton.edu
Pseudocode Yes Algorithm 1 NESTT-G Algorithm
Open Source Code No The paper does not provide any explicit statements or links indicating that open-source code for the described methodology is available.
Open Datasets No To test the performance of the proposed algorithm, we generate the problem following similar setups as [16]. Let X = (X1; , XN) RM P with P i Ni = M and each Xi RNi P corresponds to Ni data points, and it is generated from i.i.d Gaussian. This indicates the data was generated, not taken from a publicly accessible source with provided access details.
Dataset Splits No The paper does not provide specific dataset split information (percentages, sample counts, or citations to predefined splits) for training, validation, or testing.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types, or memory amounts) 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 We set M = 100,000, P = 5000, N = 50, K = 22 P,and R = ν 1. We implement NESTT-G/E, the SGD, and the nonconvex SAGA proposed in [21] with stepsize β = 1 3Lmax N2/3 (with Lmax := maxi Li).