Causal Inference via Kernel Deviance Measures

By Jovana Mitrovi'c, Dino Sejdinovic, Yee Whye Teh. 2018

68 citations recorded by the source corpus.

Discovering the causal structure among a set of variables is a fundamental problem in many areas of science. In this paper, we propose Kernel Conditional Deviance for Causal Inference (KCDC) a fully nonparametric causal discovery method based on purely observational data. From a novel interpretation of the notion of asymmetry between cause and effect, we derive a corresponding asymmetry measure using the framework of reproducing kernel Hilbert spaces. Based on this, we propose three decision rules for causal discovery. We demonstrate the wide applicability of our method across a range of diverse synthetic datasets. Furthermore, we test our method on real-world time series data and the real-world benchmark dataset Tubingen Cause-Effect Pairs where we outperform existing state-of-the-art methods.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Learning Neural Causal Models from Unknown Interventions (2019, methodology): “Asymmetry-based methods (Shimizu et al., 2006; Hoyer et al., 2009; Daniusis et al., 2012; Budhathoki & Vreeken, 2017; Mitrovic et al., 2018) assume asymmetry between cause and effect in the data and try to use this information to estimate the causal structure.”
Causal Confusion in Imitation Learning (2019, background): ““Causal discovery” approaches allow causal inference from pre-recorded observations under constraints [54, 17, 29, 15, 30, 31, 26, 14, 34, 57].”
D’ya Like DAGs? A Survey on Structure Learning and Causal Discovery (2021, background): “[252] 2020 score Granger no Bi Order-MCMC [61] 2003 sampling yes yes yes no DAG OG [54] 2008 sampling yes yes yes yes DAG EE-DAG [269] 2011 sampling yes yes yes yes DAG ZIPBN [35] 2020 sampling yes no yes no DAG LiNGAM [216] 2006 asymmetries yes no yes no DAG LV LiNGAM [103] 2008 asymmetries no yes ”
Detecting crosstalk errors in quantum information processors (2019, background): “[33, 40], and it would be interesting to study whether any of these present any advantages when post-processing lightweight experimental data for crosstalk error detection.”
Automated Machine Learning: Methods, Systems, Challenges (2019, unknown): “An example of score used in [31, 32] is the negative log-likelihood score defined for a candidate generative model with causal orientation X → Y by:”