False discovery proportion envelopes with m-consistency

Authors: Meah Iqraa, Blanchard Gilles, Roquain Etienne

JMLR 2024 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental These improvements are illustrated with numerical experiments and real data examples. In particular, the improvement is significant in the knockoffs setting, which shows the impact of the method for a practical use.
Researcher Affiliation Academia Iqraa Meah EMAIL Center for Research in Epidemiology and Statistic S (CRESS) Universit e Paris Cit e and Universit e Sorbonne Paris Nord, Inserm, INRAE F-75004 Paris, France; Gilles Blanchard EMAIL Institut Math ematiques de Orsay (IMO) CNRS, Inria, Universit e Paris-Saclay F-91405 Orsay Cedex; Etienne Roquain EMAIL Laboratoire de Probabilit es, Statistique et Mod elisation, CNRS, Sorbonne Universit e, Universit e de Paris. Sorbonne Universit e 4 place Jussieu, 75005, Paris
Pseudocode No The paper does not contain any clearly labeled pseudocode or algorithm blocks. Procedures are described mathematically or in narrative text.
Open Source Code Yes All our numerical experiments are reproducible from the code provided in the repository https://github.com/iqm15/Consistent FDP.
Open Datasets Yes We now consider the online case, by applying our method to the real data example coming from the International Mice Phenotyping Consortium (IMPC) (Mu noz-Fuentes et al., 2018)
Dataset Splits No The paper describes simulation setups and theoretical models (e.g., Gaussian location model, VCT model) for generating data for numerical experiments, but does not specify training/test/validation splits of any publicly available datasets in the machine learning sense.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., CPU, GPU models, memory) used for running the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., programming languages, libraries, frameworks) used for the experiments.
Experiment Setup Yes Throughout the experiments, the default value for δ is 0.25 and the default number of replications to evaluate each FDP bound is 1000. To investigate the consistency property, we take m varying in the range {10i, 2 i 6}, and we consider the FDP bounds FDP Simes α (16), FDP DKW α (17), FDP KR α (18), FDP Well α (19) for α {0.05, 0.1, 0.15, 0.2}.