analiza aiqbench
This paper extends distributionally robust optimization, a regularization approach previously limited to convex models, to mixed-integer sparse programs, with statistical guarantees tied to the sparsity parameter. The practical payoff is avoiding costly cross-validation for choosing sparsity in interpretable rule-based classifiers. Readers should note that the results are demonstrated on binary classification with feature-value rules, so scalability and performance on larger or less structured settings remain open questions.
Distributionally Robust Regularization of Sparse Integer Programming Trained Learning Models
arXiv:2311.01994v2 Announce Type: replace-cross Abstract: Building explainable machine learning models is crucial for human users to be able to interpret the proposed statistical relationship obtained from the training data. Mixed-integer optimization formulations are often used to train such models with explicit sparsity constraints, aiming to strike the right trade-off between sparsity and prediction accuracy. Existing methods to find the right choice of sparsity -- e.g., via cross-validation -- are computationally expensive. For convex model training formulations, recent advances in distributionally robust optimization (DRO) provide strong generalization while sidestepping this computational burden. We describe an extension of such regularization via DRO to mixed-integer sparse programs, providing statistical guarantees as a function of an associated sparsity parameter of the formulation. We illustrate the use of this approach in the case of building explainable binary classification models using sets of feature value rules.
Abstract:Building explainable machine learning models is crucial for human users to be able to interpret the proposed statistical relationship obtained from the training data. Mixed-integer optimization formulations are often used to train such models with explicit sparsity constraints, aiming to strike the right trade-off between sparsity and prediction accuracy. Existing methods to find the right choice of sparsity -- e.g., via cross-validation -- are computationally expensive. For convex model training formulations, recent advances in distributionally robust optimization (DRO) provide strong generalization while sidestepping this computational burden. We describe an extension of such regularization via DRO to mixed-integer sparse programs, providing statistical guarantees as a function of an associated sparsity parameter of the formulation. We illustrate the use of this approach in the case of building explainable binary classification models using sets of feature value rules.
Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Optimization and Control (math.OC)
arXiv:2311.01994 [stat.ML]
(or arXiv:2311.01994v2 [stat.ML] for this version)
https://doi.org/10.48550/arXiv.2311.01994 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 237 cuvinte
Subiecte conexe
Frontier Risk Report (February to March 2026)
METR · 19 mai 2026
Frontier AI Risk Report (February–March 2026)
METR · 19 mai 2026
Informe de riesgos de la IA de frontera (febrero–marzo de 2026)
METR · 19 mai 2026
Introducing North Mini Code: Cohere’s First Model For Developers
Hugging Face · 9 iun. 2026