Gaussian-Based and Outside-the-Box Runtime Monitoring Join Forces

Vahid Hashemi, Jan Křetínský, Sabine Rieder, Torsten Schön, Jan Vorhoff

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

Abstract

Since neural networks can make wrong predictions even with high confidence, monitoring their behavior at runtime is important, especially in safety-critical domains like autonomous driving. In this paper, we combine ideas from previous monitoring approaches based on observing the activation values of hidden neurons. In particular, we combine the Gaussian-based approach, which observes whether the current value of each monitored neuron is similar to typical values observed during training, and the Outside-the-Box monitor, which creates clusters of the acceptable activation values, and, thus, considers the correlations of the neurons’ values. Our experiments evaluate the achieved improvement.

OriginalspracheEnglisch
TitelRuntime Verification - 24th International Conference, RV 2024, Proceedings
Redakteure/-innenErika Ábrahám, Houssam Abbas
Herausgeber (Verlag)Springer Science and Business Media Deutschland GmbH
Seiten218-228
Seitenumfang11
ISBN (Print)9783031742330
DOIs
PublikationsstatusVeröffentlicht - 2025
Veranstaltung24th International Conference on Runtime Verification, RV 2024 - Instanbul, Türkei
Dauer: 15 Okt. 202417 Okt. 2024

Publikationsreihe

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Band15191 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Konferenz

Konferenz24th International Conference on Runtime Verification, RV 2024
Land/GebietTürkei
OrtInstanbul
Zeitraum15/10/2417/10/24

Fingerprint

Untersuchen Sie die Forschungsthemen von „Gaussian-Based and Outside-the-Box Runtime Monitoring Join Forces“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren