ASP-based Discovery of Semi-Markovian Causal Models under Weaker Assumptions
Authors: Zhalama, Jiji Zhang, Frederick Eberhardt, Wolfgang Mayer, Mark Junjie Li
IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We report two types of simulation, one using an independence oracle that specifies the true CI/CDs of the causal model, and one that uses the CI/CDs inferred from the sample data. For both simulations we followed the model generation process of [Hyttinen et al., 2014] for causally insufficient models: We generated 100 random linear Gaussian models over 6 vertices with an average edge degree of 1 for directed edges. ... Figure 3 plots the ROC curves for the inferred d-connections. ... Figure 4 shows the sorted solving times for the different background assumptions (with maximum time budget of 5,000s). |
| Researcher Affiliation | Academia | Zhalama1 , Jiji Zhang2 , Frederick Eberhardt3 , Wolfgang Mayer1 and Mark Junjie Li4 1University of South Australia 2Lingnan University 3California Institute of Technology 4Shenzhen University |
| Pseudocode | Yes | Figure 2: ASP Encoding of V-adjacency-faithfulness and V-adjacency-minimality. Inference rules for virtual-adjacency: h(X, Z, Y ) :edge(X, Z), ancestors(Z, X, Y ). h(X, Z, Y ) :conf(X, Z), ancestors(Z, X, Y ). ... |
| Open Source Code | No | The paper mentions implementing algorithms using an existing framework but does not provide a link to its own source code implementation or state that it is open source. |
| Open Datasets | No | We generated 100 random linear Gaussian models over 6 vertices with an average edge degree of 1 for directed edges. ... In the oracle setting, we randomly generated 100 linear Gaussian models with latent confounders over 6 variables and then input the independence oracles implied by these models. ... In the finite sample case we generated five data sets with 500 samples from each of the 100 models. |
| Dataset Splits | No | The paper describes generating models and data sets but does not specify a training/validation/test split or cross-validation methodology. |
| Hardware Specification | No | The paper does not specify any hardware details such as GPU/CPU models, memory, or cloud instance types used for running experiments. |
| Software Dependencies | No | The paper mentions using 'Answer Set Programming (ASP)' and 'off-the-shelf solvers' within the framework of [Hyttinen et al., 2014], but it does not specify any version numbers for these software components or libraries. |
| Experiment Setup | Yes | The edge coefficients were drawn uniformly from [ 0.8, 0.2] [0.2, 0.8]. The error covariance matrices (which also represent the confounding) were generated using the observational covariance matrix of a similarly constructed causally sufficient model (with its error covariances sampled from N(0.5, 0.01)). ... We used correlational t-tests and tried 10 threshold values for rejecting the null hypothesis (0.0001, 0.0005, 0.001, 0.005, 0.01, 0.05, 0.1, 0.15, 0.2, 0.25). The test results formed the input for the algorithms. We also used the log-weighting scheme and tried 10 values for the free parameter of the Bayesian test (0.05, 0.09, 0.1, 0.15, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.9). |