Pick-to-Learn: state-of-the-art safety guarantees for machine learning and control

AI models are increasingly embedded in scientific research and industrial production, where they inform predictions and drive decisions. Yet, in safety-critical settings including autonomous driving and medical diagnostics, deploying these models requires rigorous safety and performance guarantees — a need that has motivated much recent work at the interface of statistical learning, optimization, and control. However, existing approaches, including conformal prediction, test-set methods, and PAC-Bayes bounds, face two major limitations: they either require setting aside part of the dataset to generate guarantees — possibly degrading the quality of the learned model —  or they produce bounds that often do not reflect true performance.
In this talk, I will present a recent line of work that overcomes these limitations by enabling the use of all available data to jointly train models and equip them with tight safety or performance guarantees. The core technical idea is to embed any black-box learner into a suitably constructed meta-algorithm, Pick-to-Learn, which turns the original black-box algorithm into a sample compression scheme from which sharp guarantees can be derived.I will then show how, across a breadth of applications in machine learning and data-driven control, P2L delivers both better-performing models and tighter certificates than the state of the art, highlighting its potential for broad practical impact.    
 

Join at imt.lu/aula2