• English
  • Deutsch
  • Log In
    Password Login
    Have you forgotten your password?
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Artikel
  4. The Good and the Bad: Using Neuron Coverage as a DNN Validation Technique
 
  • Details
  • Full
Options
June 18, 2022
Book Article
Title

The Good and the Bad: Using Neuron Coverage as a DNN Validation Technique

Abstract
Verification and validation (V&V) is a crucial step for the certification and deployment of deep neural networks (DNNs). Neuron coverage, inspired by code coverage in software testing, has been proposed as one such V&V method. We provide a summary of different neuron coverage variants and their inspiration from traditional software engineering V&V methods. Our first experiment shows that novelty and granularity are important considerations when assessing a coverage metric. Building on these observations, we provide an illustrative example for studying the advantages of pairwise coverage over simple neuron coverage. Finally, we show that there is an upper bound of realizable neuron coverage when test data are sampled from inside the operational design domain (in-ODD) instead of the entire input space.
Author(s)
Gannamaneni, Sujan Sai  
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS  
Akila, Maram  
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS  
Heinzemann, Christian  
Robert Bosch GmbH
Woehrle, Matthias  
Robert Bosch GmbH
Mainwork
Deep Neural Networks and Data for Automated Driving  
Project(s)
Methoden und Maßnahmen zur Absicherung von KI basierten Wahrnehmungsfunktionen für das automatisierte Fahren; Teilvorhaben: Entwicklung und Test von Absicherungsmethoden, Ausreißeridentifikation, logische Methoden und visuelle Analytic, Anbindung an die Wissenschaft  
Funder
Bundesministerium für Wirtschaft und Klimaschutz  
Open Access
DOI
10.1007/978-3-031-01233-4_14
10.24406/publica-4145
File(s)
Gannamaneni_Neuron_Coverage_2022.pdf (589.46 KB)
Rights
CC BY 4.0: Creative Commons Attribution
Language
English
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS  
Keyword(s)
  • Neuron Coverage

  • DNN Testing

  • Robustness

  • Operational Design Domains (ODDs)

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024