Weakly Submodular Maximization Beyond Cardinality Constraints: Does Randomization Help Greedy?

Authors: Lin Chen, Moran Feldman, Amin Karbasi

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

Reproducibility Variable Result LLM Response
Research Type Experimental Moreover, our experimental results show that our proposed algorithm performs well in a variety of real-world problems, including regression, video summarization, splice site detection, and black-box interpretation.
Researcher Affiliation Academia 1Yale Institute for Network Science, Yale University, New Haven, CT, USA 2Department of Electrical Engineering, Yale University 3 Department of Mathematics and Computer Science, Open University of Israel, Ra anana, Israel.
Pseudocode Yes Algorithm 1 Residual Random Greedy for Matroids
Open Source Code No The paper does not provide explicit access (link, statement of release) to its own source code.
Open Datasets Yes A detailed description of this dataset is presented in (Yeo & Burge, 2004).
Dataset Splits No The paper describes the characteristics and generation of datasets used (e.g., n=100, p=200 for synthetic data; MEMset dataset details) but does not specify explicit training, validation, or test splits (e.g., 80/10/10%).
Hardware Specification No The paper does not specify any particular hardware components (e.g., CPU, GPU models) used for running the experiments.
Software Dependencies No The paper mentions software like "logistic regression", "LIME framework", and "SLIC algorithm" but does not provide specific version numbers for any of these or other dependencies.
Experiment Setup Yes We chose n = 100 and p = 200, and constructed each row of the n p matrix X independently according to an autoregressive (AR) process with α = 0.5 and noise variance σ2 = 10.