On the Complexity of Counterfactual Reasoning

Authors: Yunqiu Han, Yizuo Chen, Adnan Darwiche

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We finally present empirical results in Section 7 which reveal that, on average, the complexity gap between counterfactual and associational/interventional reasoning on fully specified SCMs can be smaller than what our worst-case bounds may suggest. 7 Experimental Results We consider experiments that target random networks whose structures emulate the structures of SCMs used in counterfactual reasoning.
Researcher Affiliation Academia Yunqiu Han , Yizuo Chen , Adnan Darwiche University of California, Los Angeles yunqiu21@g.ucla.edu, yizuo.chen@ucla.edu, darwiche@cs.ucla.edu
Pseudocode Yes Algorithm 1 Jointree to Twin Jointree
Open Source Code No No explicit statement providing access to the source code for the methodology described in this paper was found.
Open Datasets No The paper describes methods for generating random networks ('We generated random networks according to the method used in [Darwiche, 2020].' and 'We also included experiments using random networks generated according to the method in [Ide and Cozman, 2002].'), but does not provide concrete access information (link, DOI, repository, or formal citation with authors/year) for a publicly available or open dataset used for training or evaluation.
Dataset Splits No The paper describes generating random networks and evaluating their properties, but does not mention specific training, validation, or test dataset splits.
Hardware Specification No No specific hardware details (GPU/CPU models, memory, or detailed computer specifications) used for running experiments were mentioned in the paper.
Software Dependencies No No specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment were provided.
Experiment Setup Yes We used n {50, 75, 100, 125, 150, 200, 250, 300} and p {3, 5, 7}. For each combination of n and p, we generated 50 random, base networks and reported averages of two properties for the constructed jointrees: width and normalized width. We use the popular minfill heuristic [Kjaerulff, 1990] in our experiments.