Randomized Experimental Design for Causal Graph Discovery

Authors: Huining Hu, Zhentao Li, Adrian R Vetta

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we present computer simulations to complement our theoretic results.
Researcher Affiliation Academia Huining Hu School of Computer Science, Mc Gill University. huining.hu@mail.mcgill.ca Zhentao Li LIENS, Ecole Normale Sup erieure zhentao.li@ens.fr Adrian Vetta Department of Mathematics and Statistics and School of Computer Science, Mc Gill University. vetta@math.mcgill.ca
Pseudocode No The paper does not contain any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper does not provide any statement or link regarding the availability of its source code.
Open Datasets No The paper describes generating its own random causal graphs using the Erd os-R enyi model ("For the simulations, we first generate a random causal graph G in the E-R model."), rather than using a pre-existing publicly available dataset with concrete access information.
Dataset Splits No The paper describes simulation parameters for generating graphs but does not specify training, validation, or test dataset splits.
Hardware Specification No The paper mentions that simulations were conducted in MATLAB but does not specify any hardware details such as CPU, GPU, or memory used.
Software Dependencies No The paper mentions that simulations were conducted in MATLAB, but does not specify a version number for MATLAB or any other software dependencies.
Experiment Setup Yes For the simulations, we first generate a random causal graph G in the E-R model. We ran simulations for four choices of probability p, specifically p {0.8, 0.5, 0.1, 0.01}, and for four choices of graph size n, specifically n {500, 1000, 5000, 15000}. For each combination pair {n, p} we ran 1000 simulations.