Causal Explanation Under Indeterminism: A Sampling Approach

Authors: Christopher Merck, Samantha Kleinberg

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In contrast we make two main contributions: an algorithm for explanation that calculates the strength of token causes, and an evaluation based on simulated data that enables objective comparison against prior methods and ground truth. We show that the approach finds the correct relationships in classic test cases (causal chains, common cause, and backup causation) and in a realistic scenario (explaining hyperglycemic episodes in a simulation of type 1 diabetes).
Researcher Affiliation Academia Christopher A Merck and Samantha Kleinberg Stevens Institute of Technology Hoboken NJ
Pseudocode Yes Algorithm 1 Causal Explanation in Stochastic Processes
Open Source Code Yes 2https://github.com/kleinberg-lab/stoch_cf
Open Datasets No The paper uses simulated data based on models (e.g., 'simulation of the human glucose-insulin system') but does not provide access information (link, DOI, citation) to a publicly available dataset used for training or general use.
Dataset Splits No The paper does not specify training, validation, or test dataset splits. It describes using '1000 samples' for evaluation within simulations, which is not a dataset split.
Hardware Specification No The paper only mentions 'All simulations in the paper took < 60sec on a workstation' which is not a specific hardware specification.
Software Dependencies No The paper mentions 'a python implementation' but does not specify version numbers for Python or any specific libraries/packages used.
Experiment Setup Yes We apply Algorithm 1 to three test cases using 1000 samples and a time horizon of 20 time steps. ... We simulate a 60 kg patient with T1DM wearing an insulin pump delivering a basal insulin infusion of 3.5 pmol/kg/min. The patient goes for a run R at 7 AM whereby heart rate is increased from 60 bpm to 90 bpm for 30 minutes and a snack containing 27.5 g of glucose is consumed to offset the glucose consumed by the exercise. ... using 1000 samples and a time horizon of 15 minutes.