Telling cause from effect in deterministic linear dynamical systems

Authors: Naji Shajarisales, Dominik Janzing, Bernhard Schoelkopf, Michel Besserve

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments show encouraging results on synthetic as well as real-world data.
Researcher Affiliation Academia 1 MPI for Intelligent Systems, Tuebingen, Germany 2 MPI for Biological Cybernetics , Tuebingen, Germany
Pseudocode Yes Algorithm 1 SIC Inference
Open Source Code No The paper does not provide an explicit statement about releasing its own source code for the methodology, nor does it provide a direct link to a code repository for its implementation. The mentioned links refer to datasets or third-party libraries.
Open Datasets Yes To do a comparison with Granger causality, we applied our framework to recordings from those regions using a publicly available dataset1 (Mizuseki et al., 2009; 2006). 1http://crcns.org/data-sets/hc
Dataset Splits No The paper describes generating synthetic data and processing real-world time series by dividing them into intervals, but it does not specify explicit training, validation, and test dataset splits with proportions or sample counts.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions using 'statsmodel Python library' but does not provide specific version numbers for Python, statsmodels, or any other ancillary software components.
Experiment Setup Yes We simulated sequences of length 1000. The PSD of X and Y were estimated using Welch s method (Welch, 1967). (Section 4.1); We divided the duration of ten minutes into 300 intervals of two seconds (N = 2504) to reduce the effect of nonstationarity in data analysis, and performed SIC causal inference on each interval for each electrode pair. (Section 4.2.3)