Skip to Main content Skip to Navigation
Conference papers

Overestimation learning with guarantees

Abstract : We describe a complete method that learns a neural network which is guaranteed to overestimate a reference function on a given domain. The neural network can then be used as a surrogate for the reference function. The method involves two steps. In the first step, we construct an adaptive set of Majoring Points. In the second step, we optimize a well-chosen neural network to overestimate the Majoring Points. In order to extend the guarantee on the Majoring Points to the whole domain, we necessarily have to make an assumption on the reference function. In this study, we assume that the reference function is monotonic. We provide experiments on synthetic and real problems. The experiments show that the density of the Majoring Points concentrate where the reference function varies. The learned over-estimations are both guaranteed to overestimate the reference function and are proven empirically to provide good approximations of it. Experiments on real data show that the method makes it possible to use the surrogate function in embedded systems for which an underestimation is critical; when computing the reference function requires too many resources.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03118707
Contributor : Francois Malgouyres <>
Submitted on : Monday, January 25, 2021 - 11:32:43 AM
Last modification on : Wednesday, June 9, 2021 - 10:00:10 AM
Long-term archiving on: : Monday, April 26, 2021 - 6:26:28 PM

Files

safelearning.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03118707, version 1
  • ARXIV : 2101.11717

Citation

Adrien Gauffriau, François Malgouyres, Mélanie Ducoffe. Overestimation learning with guarantees. AAAI-21, workshop on safeAI, Feb 2021, Valence (Virtual), Spain. ⟨hal-03118707⟩

Share

Metrics

Record views

123

Files downloads

68