Global Convergence of Least Squares EM for Demixing Two Log-Concave Densities

Authors: Wei Qian, Yuqian Zhang, Yudong Chen

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Moreover, empirical results suggest that the log-convexity assumption cannot be relaxed completely: Figure 1 provides an example where the LS-EM algorithm may converge to 0 (an undesired solution) with constant probability when the log-concavity property is violated. See Appendix H for additional numerical results.
Researcher Affiliation Academia Wei Qian, Yuqian Zhang, Yudong Chen School of Operations Research and Information Engineering Cornell University
Pseudocode No The paper describes the LS-EM algorithm steps in text in Section 3 but does not provide a formal algorithm box or pseudocode block.
Open Source Code No No statement regarding the release of open-source code for the described methodology was found.
Open Datasets No The paper analyzes the algorithm under finite sample settings but does not refer to the use of a specific, publicly available dataset with concrete access information.
Dataset Splits No The paper does not provide specific details on dataset splits (training, validation, test) as it focuses on theoretical analysis, with limited empirical results not detailing data splits.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory) used for running experiments were mentioned.
Software Dependencies No No specific software dependencies with version numbers were mentioned for replicating the experiments.
Experiment Setup No While numerical experiments are mentioned in the paper, specific experimental setup details such as hyperparameter values, training configurations, or system-level settings are not provided in the main text.