Causal Discovery from Multiple Data Sets with Non-Identical Variable Sets

Authors: Biwei Huang, Kun Zhang, Mingming Gong, Clark Glymour10153-10161

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on various synthetic and real-world data sets are presented to demonstrate the efficacy of our methods. 7 Experimental Results To show the efficacy of the proposed approach for causal discovery from non-identical data sets, we apply it to both synthetic and real-world data.
Researcher Affiliation Academia 1Department of Philosophy, Carnegie Mellon University, Pittsburgh, PA, USA 2School of Mathematics and Statistics, University of Melbourne, Melbourne, Australia
Pseudocode Yes Algorithm 1 Adversarial Learning-Based Estimation of the Causal Adjacency Matrix
Open Source Code No The paper does not provide an explicit statement or link for open-source code availability for the described methodology.
Open Datasets Yes We applied CD-Mi Ni to multivariate flow cytometry data, which were measured from 11 phosphorylated proteins and phospholipids (Sachs et al. 2005).
Dataset Splits No The paper does not explicitly provide details about standard training, validation, or test dataset splits (e.g., percentages or counts for each split).
Hardware Specification No The paper does not explicitly describe the specific hardware (e.g., CPU, GPU models, memory) used to run its experiments.
Software Dependencies No The paper mentions various algorithms and methods but does not provide specific software dependencies (e.g., library names with version numbers) needed to replicate the experiments.
Experiment Setup Yes Each noise term ei is modeled with a mixture of two Gaussian components, with mean μi,k U( 0.6, 0.3) U(0.3, 0.6), variance σ2 i,k U(0.1, 0.5), and the mixture proportion πi,k U(0.3, 0.6) with 2 k=1 πi,k = 1. The non-zero entries of the causal adjacency matrix B was generated according to bij U( 0.8, 0.3) U(0.3, 0.8). For the results from the proposed methods, the final graph is determined by setting a threshold on the estimated causal adjacency matrix ˆB; we used 0.1 as the threshold, that is, the estimated graph ˆGij = 1 if |ˆbij| > 0.1, and ˆGij = 0 if otherwise.