Multilabel Classification with Group Testing and Codes

Authors: Shashanka Ubaru, Arya Mazumdar

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

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical experiments with various datasets illustrate the superior performance of our method.In this section, we illustrate the performance of the proposed group testing approach in the multilabel classification problems (MLGT) via several numerical experiments on various datasets.
Researcher Affiliation Academia 1Department of Computer Science and Engineering, University of Minnesota at Twin Cities, MN USA. 2College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA..
Pseudocode Yes Algorithm 1 MLGT: Training Algorithm and Algorithm 2 MLGT: Prediction Algorithm
Open Source Code No The paper thanks 'Dr. Manik Varma and his team for making many MLC datasets and codes available online.' but does not provide a link or explicit statement for the open-source code of the methodology presented in this paper.
Open Datasets Yes Datasets: We use some popular publicly available multilabel datasets in our experiments. All datasets were obtained from The Extreme Classification Repository1 (Bhatia et al., 2015). The footnote directs to "https://manikvarma.github.io/downloads/XC/XMLRepository.html".
Dataset Splits No The paper specifies 'training points n' and 'test points nt' for the datasets, as seen in Table 1 (e.g., 'n = 2000, nt = 501'), but does not explicitly mention a distinct validation split or set.
Hardware Specification No The paper states 'The runtimes reported (using cputime in Matlab)' but does not specify any hardware details such as CPU, GPU models, or memory.
Software Dependencies No The paper mentions 'Matlab' and the use of 'Least squares regression' and 'Orthogonal Matching Pursuit (OMP)', but does not provide specific version numbers for any software dependencies.
Experiment Setup No The paper specifies the use of 'Least squares binary classifier' for MLGT and 'Least squares regression with ℓ2 regularization (ridge regression)' and 'Orthogonal Matching Pursuit (OMP)' for MLCS, but does not provide specific experimental setup details such as hyperparameters (e.g., learning rate, batch size, number of epochs) or optimizer settings.