DYNOTEARS: Structure Learning from Time-Series Data
This work addresses the challenge of scalable and accurate structure learning for dynamic Bayesian networks, which is incremental as it builds on existing score-based methods with a novel acyclicity constraint.
The authors tackled the problem of learning dynamic Bayesian network structures from time-series data by proposing DYNOTEARS, a score-based method that simultaneously estimates contemporaneous and time-lagged relationships with an acyclicity constraint, which outperformed other methods on simulated data, especially in high dimensions, and showed scalability and accuracy on real datasets in finance and molecular biology.
We revisit the structure learning problem for dynamic Bayesian networks and propose a method that simultaneously estimates contemporaneous (intra-slice) and time-lagged (inter-slice) relationships between variables in a time-series. Our approach is score-based, and revolves around minimizing a penalized loss subject to an acyclicity constraint. To solve this problem, we leverage a recent algebraic result characterizing the acyclicity constraint as a smooth equality constraint. The resulting algorithm, which we call DYNOTEARS, outperforms other methods on simulated data, especially in high-dimensions as the number of variables increases. We also apply this algorithm on real datasets from two different domains, finance and molecular biology, and analyze the resulting output. Compared to state-of-the-art methods for learning dynamic Bayesian networks, our method is both scalable and accurate on real data. The simple formulation and competitive performance of our method make it suitable for a variety of problems where one seeks to learn connections between variables across time.