CAnDOIT: Causal Discovery with Observational and Interventional Data from Time-Series

TL;DR

CAnDOIT is a novel causal discovery method that integrates observational and interventional time-series data to accurately reconstruct causal models, especially in complex, real-world scenarios like robotics.

Contribution

The paper introduces CAnDOIT, a new method that effectively combines observational and interventional data for causal discovery in time-series, improving accuracy in complex environments.

Findings

Successfully reconstructs causal models from synthetic data.

Effectively utilizes interventional data to improve causal inference.

Demonstrates robustness in robotic manipulation benchmarks.

Abstract

The study of cause-and-effect is of the utmost importance in many branches of science, but also for many practical applications of intelligent systems. In particular, identifying causal relationships in situations that include hidden factors is a major challenge for methods that rely solely on observational data for building causal models. This paper proposes CAnDOIT, a causal discovery method to reconstruct causal models using both observational and interventional time-series data. The use of interventional data in the causal analysis is crucial for real-world applications, such as robotics, where the scenario is highly complex and observational data alone are often insufficient to uncover the correct causal structure. Validation of the method is performed initially on randomly generated synthetic models and subsequently on a well-known benchmark for causal structure learning in a robotic manipulation environment. The experiments demonstrate that the approach can effectively handle data from interventions and exploit them to enhance the accuracy of the causal analysis. A Python implementation of CAnDOIT has also been developed and is publicly available on GitHub: https://github.com/lcastri/causalflow.