Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

Model-Preserving Sensitivity Analysis for Families of Gaussian Distributions

Authors: Christiane Görgen, Manuele Leonelli

JMLR 2020 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our methods are demonstrated to be robust and comparable to standard ones, which can break the conditional independence structure of the model, using an artificial example and a medical real-world application.
Researcher Affiliation Academia Christiane G orgen EMAIL Max Planck Institute for Mathematics in the Sciences Inselstraße 22, 04103 Leipzig, Germany Manuele Leonelli EMAIL School of Human Sciences and Technology IE University Madrid, Spain
Pseudocode No The paper describes methods and proofs using mathematical notation but does not include any explicitly labeled pseudocode or algorithm blocks with structured steps.
Open Source Code Yes An implementation of the methods developed in this paper in the open-source R software (R Core Team, 2019) is given in the package bnmonitor and available at https://github. com/manueleleonelli/bnmonitor.
Open Datasets Yes In this section we study a subset of the data set of Eisner et al. (2011) including metabolomic information of 77 individuals: 47 of them suffering of cachexia, whilst the remaining do not.
Dataset Splits No The paper mentions analyzing a dataset split into two populations (ill and not ill) and generating random data for other networks, but it does not provide specific percentages, sample counts, or methodologies for train/test/validation splits for model evaluation.
Hardware Specification Yes Computations were carried out on a Intel Core I7 of 8th generation.
Software Dependencies No The paper mentions using "open-source R software (R Core Team, 2019)" and the "bnlearn R package (Scutari, 2010)", but it does not specify exact version numbers for these software components or any other key libraries.
Experiment Setup No For the artificial example, the paper sets specific model parameters (e.g., β0i = 0, vi = 1, β12 = 2) to construct a covariance matrix. For the real-world application, it states that GBN models were