Abstract : In order to reach satisfactory performance, fault diagnosis methods require the tuning of internal parameters, usually called hyperparameters. This is generally achieved by optimizing a performance criterion, typically a trade-off between false-alarm and non-detection rates. Perturbations should also be taken into account, for instance by considering the worst possible case. A new method to achieve such a tuning is described, which is especially interesting when the simulations required are so costly that their number is severely limited. It achieves min-max optimization of the tuning parameters via a relaxation procedure and Kriging-based optimization. This approach is applied to the worst-case optimal tuning of a fault diagnosis method consisting of an observer-based residual generator followed by a statistical test. It readily extends to the tuning of hyperparameters in other contexts.
https://hal-supelec.archives-ouvertes.fr/hal-00615618
Contributor : Julien Marzat
<>
Submitted on : Thursday, September 8, 2011 - 9:22:35 AM
Last modification on : Tuesday, March 26, 2019 - 2:24:42 PM
Document(s) archivé(s) le : Friday, December 9, 2011 - 2:20:29 AM
Julien Marzat, Hélène Piet-Lahanier, Eric Walter. Min-max hyperparameter tuning, with application to fault detection. 18th IFAC World Congress, Aug 2011, Milan, Italy. pp.12904-12909. ⟨hal-00615618⟩
