Causal Effect Identification in LiNGAM Models with Latent Confounders

Authors: Daniele Tramontano, Yaroslav Kivva, Saber Salehkaleybar, Mathias Drton, Negar Kiyavash

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 5. Experimental Results, We used Algorithms 1 and 2 to check the identifiability of a causal effect for randomly selected edges in random graphs. Experimental results show the effectiveness of the proposed method in estimating the causal effects.
Researcher Affiliation Academia 1Technical University of Munich, Munich, Germany 2Ecole Polytechnique F ed erale de Lausanne, Lausanne, Switzerland 3Leiden Institute of Advanced Computer Science, Leiden University, Netherlands 4Munich Center for Machine Learning, Munich, Germany.
Pseudocode Yes Algorithm 1 Total Causal Effect Identification with Known Graph INPUT: V = O L, G, {ch(i) | i V}, (j, i) 1: ID TRUE 2: Sort V according to an ascending topological order...
Open Source Code Yes 3The code to replicate the experiments can be found at : https://github.com/danieletramontano/Causal-Effect Identification-in-Li NGAM-Models-with-Latent-Confounders.
Open Datasets No All the experiments in this subsection are done on the synthetic data generated according to the specific causal structure established for it. To generate synthetic data we specify all exogenous noises to be i.i.d., and select all non-zero entries within the matrix A through uniform sampling from [ 1, 0.5] [0.5, 1].
Dataset Splits No No specific dataset split information (percentages, sample counts, citations to predefined splits, or detailed splitting methodology) was found.
Hardware Specification No No specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) were provided.
Software Dependencies No No specific ancillary software details (e.g., library or solver names with version numbers) were provided.
Experiment Setup Yes All exogenous noises are i.i.d., and select all non-zero entries within the matrix A through uniform sampling from [ 1, 0.5] [0.5, 1]. All causal coefficients in both settings are set to one. We assumed all exogenous noises have the same distribution.