Pick-to-Learn for Systems and Control: Data-driven Synthesis with State-of-the-art Safety Guarantees

2025-12-04

AI Summary

Data-driven control for safety-critical systems faces fundamental bottlenecks: formal safety verification typically requires reserved calibration or validation datasets, restricts admissible learning algorithms, and compromises control performance. Method: This paper proposes Pick-to-Learn (P2L), the first framework enabling joint optimization of control design and formal safety verification. P2L eliminates the need for held-out data, supports arbitrary learning algorithms, and unifies statistical learning, optimal control, reachability analysis, and robust control theory into a single, end-to-end data-driven design–verification loop. Contribution/Results: Under rigorous theoretical guarantees, P2L simultaneously improves both closed-loop performance and formal safety assurance across multiple benchmark tasks. It achieves 100% data utilization—outperforming state-of-the-art methods—and establishes a scalable, verifiable paradigm for high-assurance autonomous systems.

Technology Category

• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Data-driven methods have become paramount in modern systems and control problems characterized by growing levels of complexity. In safety-critical environments, deploying these methods requires rigorous guarantees, a need that has motivated much recent work at the interface of statistical learning and control. However, many existing approaches achieve this goal at the cost of sacrificing valuable data for testing and calibration, or by constraining the choice of learning algorithm, thus leading to suboptimal performances. In this paper, we describe Pick-to-Learn (P2L) for Systems and Control, a framework that allows any data-driven control method to be equipped with state-of-the-art safety and performance guarantees. P2L enables the use of all available data to jointly synthesize and certify the design, eliminating the need to set aside data for calibration or validation purposes. In presenting a comprehensive version of P2L for systems and control, this paper demonstrates its effectiveness across a range of core problems, including optimal control, reachability analysis, safe synthesis, and robust control. In many of these applications, P2L delivers designs and certificates that outperform commonly employed methods, and shows strong potential for broad applicability in diverse practical settings.

Problem

• Ensuring safety in data-driven control without sacrificing data for testing
• Providing performance guarantees without constraining learning algorithm choices
• Applying framework to core control problems like optimal and robust control

Innovation

• Framework enables data-driven control with safety guarantees
• Uses all data for synthesis without separate calibration sets
• Outperforms existing methods in optimal and robust control