Throwing Vines at the Wall: Structure Learning via Random Search
This work addresses a key bottleneck in multivariate dependence modeling for machine learning applications, offering incremental improvements over existing heuristics.
The paper tackled the challenge of structure learning in vine copulas by proposing random search algorithms and a statistical framework with theoretical guarantees, resulting in methods that consistently outperform state-of-the-art approaches on real-world datasets.
Vine copulas offer flexible multivariate dependence modeling and have become widely used in machine learning, yet structure learning remains a key challenge. Early heuristics like the greedy algorithm of Dissmann are still considered the gold standard, but often suboptimal. We propose random search algorithms that improve structure selection and a statistical framework based on model confidence sets, which provides theoretical guarantees on selection probabilities and a powerful foundation for ensembling. Empirical results on several real-world data sets show that our methods consistently outperform state-of-the-art approaches.