Throwing Vines at the Wall: Structure Learning via Random Search
Not on the map. Why?
Abstract (arXiv)
Vine copulas offer flexible multivariate dependence modeling and have become widely used in machine learning. Yet, structure learning remains a key challenge. Early heuristics, such as Dissmann's greedy algorithm, are still considered the gold standard but are often suboptimal. We propose random search algorithms and a statistical framework based on model confidence sets, to improve structure selection, provide theoretical guarantees on selection probabilities and excess risk, as well as serve as a foundation for ensembling. Empirical results on real-world data sets show that our methods consistently outperform state-of-the-art approaches.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 23 of its 78 references that are in the map. First posted on arXiv 22 Oct 2025.
Builds on
23 of 78 references.
References · 23
- 2019Copulas as High-Dimensional Generative Models: Vine Copula Autoencoderscited by 20
- 2023BOtied: Multi-objective Bayesian optimization with tied multivariate rankscited by 14
- 2017Nonparametric estimation of simplified vine copula models: comparison of methodscited by 13
- 2017Cross-Validation with Confidencecited by 13
- 2024Semiparametric conformal predictioncited by 13
- 2018Generative Models for Simulating Mobility Trajectoriescited by 10
- 2018Prediction based on conditional distributions of vine copulascited by 8
- 2018Learning Vine Copula Models For Synthetic Data Generationcited by 8
- 2024Winners with Confidence: Discrete Argmin Inference with an Application to Model Selectioncited by 7
- 2016From Dependence to Causationcited by 6
- 2018Model selection in sparse high-dimensional vine copula models with application to portfolio riskcited by 6
- 2015D-vine copula based quantile regressioncited by 5
11 more
- 2021Nonparametric C- and D-vine based quantile regressioncited by 5
- 2024Simplified vine copula models: state of science and affairscited by 5
- 2017D-vine quantile regression with discrete variablescited by 4
- 2018Solving estimating equations with copulascited by 4
- 2015Sequential Bayesian Model Selection of Regular Vine Copulascited by 3
- 2017Selection of Sparse Vine Copulas in High Dimensions with the Lassocited by 3
- 2016Representing sparse Gaussian DAGs as sparse R-vines allowing for non-Gaussian dependencecited by 2
- 2017Dependence Modeling in Ultra High Dimensions with Vine Copulas and the Graphical Lassocited by 2
- 2022Black-Box Model Confidence Sets Using Cross-Validation with High-Dimensional Gaussian Comparisoncited by 2
- 2017Growing simplified vine copula trees: improving Dissmann's algorithmcited by 1
- 2024Towards more realistic climate model outputs: A multivariate bias correction based on zero-inflated vine copulascited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.