Learning Why Things Change: The Difference-Based Causality Learner
This addresses the challenge of causal discovery in dynamic systems for researchers in fields like neuroscience and time series analysis, offering a novel method but with incremental improvements over existing approaches.
The paper tackles the problem of learning causal models from time series data by introducing the Difference-Based Causality Learner (DBCL), which uses difference equations to represent causation and proves correctness for structure learning, including feedback loop identification, and shows empirical advantages over methods like VAR and Granger causality, with an application to EEG data for discovering causal directions in alpha rhythms.
In this paper, we present the Difference- Based Causality Learner (DBCL), an algorithm for learning a class of discrete-time dynamic models that represents all causation across time by means of difference equations driving change in a system. We motivate this representation with real-world mechanical systems and prove DBCL's correctness for learning structure from time series data, an endeavour that is complicated by the existence of latent derivatives that have to be detected. We also prove that, under common assumptions for causal discovery, DBCL will identify the presence or absence of feedback loops, making the model more useful for predicting the effects of manipulating variables when the system is in equilibrium. We argue analytically and show empirically the advantages of DBCL over vector autoregression (VAR) and Granger causality models as well as modified forms of Bayesian and constraintbased structure discovery algorithms. Finally, we show that our algorithm can discover causal directions of alpha rhythms in human brains from EEG data.