<ul><li>Algorithmic stability is a central notion in learning theory that quantifies the sensitivity of an algorithm to small changes in the training data.</li><li>Recent results establish that testing the stability of a black-box algorithm is impossible, given limited data from an unknown distribution.</li><li>This work examines the hardness of testing algorithmic stability in a broad range of settings, including categorical data.</li><li>The study finds that if the available data is limited, exhaustive search is essentially the only universally valid mechanism for certifying algorithmic stability, implying fundamental limits on stability testing.</li></ul>

Is Algorithmic Stability Testable? A Unified Framework under Computational Constraints

Discover more