Accelerating Greedy Coordinate Descent Methods

Authors: Haihao Lu, Robert Freund, Vahab Mirrokni

ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In Section 5, we present results of our numerical experiments using AGCD and ASCD on synthetic linear regression problems as well as practical logistic regression problems.
Researcher Affiliation Collaboration 1Department of Mathematics and Operations Research Center, MIT 2Sloan School of Management, MIT 3Google Research.
Pseudocode Yes Algorithm 1 Accelerated Coordinate Descent Framework without Strong Convexity
Open Source Code No The paper does not provide an explicit statement or link indicating that the source code for the methodology is openly available.
Open Datasets Yes Using this strategy, we solved a large number of logistic regression instances from LIBSVM (Chang & Lin, 2011).
Dataset Splits No The paper uses datasets like 'synthetic linear regression problems' and 'logistic regression instances from LIBSVM' but does not specify the train/validation/test splits, percentages, or sample counts used for experiments.
Hardware Specification No The paper does not explicitly describe the specific hardware (e.g., GPU/CPU models, memory) used to run its experiments.
Software Dependencies No The paper mentions using LIBSVM but does not provide specific version numbers for it or any other software dependencies.
Experiment Setup No The paper states that 'algorithm implementation details are described in the supplementary materials', but the main text does not include specific hyperparameters, optimizer settings, or other detailed experimental setup information.