Conditional distribution variability measures for causality detection
This work addresses causality detection, a foundational problem in statistics and machine learning, with incremental improvements in performance for specific benchmark tasks.
The authors tackled the problem of inferring causal-effect relationships by deriving variability measures for conditional probability distributions and combining them with standard statistical measures. Their model achieved an AUC score of 0.82 on the final test database, ranking second in the ChaLearn cause-effect pair challenge.
In this paper we derive variability measures for the conditional probability distributions of a pair of random variables, and we study its application in the inference of causal-effect relationships. We also study the combination of the proposed measures with standard statistical measures in the the framework of the ChaLearn cause-effect pair challenge. The developed model obtains an AUC score of 0.82 on the final test database and ranked second in the challenge.