Experimental Design on a Budget for Sparse Linear Models and Applications

Authors: Sathya Narayanan Ravi, Vamsi Ithapu, Sterling Johnson, Vikas Singh

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

Reproducibility Variable Result LLM Response
Research Type Experimental We perform a detailed set of experiments, on benchmarks and a large neuroimaging study, showing that the proposed models are effective in practice.
Researcher Affiliation Academia Sathya N. Ravi RAVI5@WISC.EDU Vamsi K. Ithapu ITHAPU@WISC.EDU Sterling C. Johnson SCJ@MEDICINE.WISC.EDU Vikas Singh VSINGH@BIOSTAT.WISC.EDU University of Wisconsin Madison William S. Middleton Memorial Veterans Hospital
Pseudocode Yes Algorithm 1 Alternating Minimization Algorithm, Algorithm 2 Randomized coordinate descent algorithm for solving (12)
Open Source Code Yes The code is publicly available at https://github.com/sraviuwmadison/Expdesign_sparse.
Open Datasets Yes two standard LASSO datasets (prostate, (Tibshirani, 1996) and lars, (Efron et al., 2004)) and Alzheimer s Disease Neuroimaging Initiative (ADNI) (neuro).
Dataset Splits No The paper refers to 'full model' and 'reduced setup' comparisons, and mentions 'train' and 'test' data in the context of the Alzheimer's study, but does not provide specific train/validation/test split percentages or sample counts for any dataset.
Hardware Specification Yes A single workstation with 8 cores and 32GB RAM is used for experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes We ran 1000 epochs of EDI, and 50 main iterations of ED-S (with 20 iterations for each of its subproblems).