NEUROSPF: A Tool for the Symbolic Analysis of Neural Networks

This paper presents NEUROSPF, a tool for the symbolic analysis of neural networks. Given a trained neural network model, the tool extracts the architecture and model parameters and translates them into a Java representation that is amenable for analysis using the Symbolic PathFinder symbolic executi...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2021 IEEE/ACM 43rd International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) s. 25 - 28
Hlavní autoři: Usman, Muhammad, Noller, Yannic, Pasareanu, Corina S., Sun, Youcheng, Gopinath, Divya
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.05.2021
Témata:
Analytical models
Maintenance engineering
Neural networks
Software engineering
Symbolic Execution
Symbolic PathFinder
Testing
Uniform resource locators
ISBN:1665412194, 9781665412193
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper presents NEUROSPF, a tool for the symbolic analysis of neural networks. Given a trained neural network model, the tool extracts the architecture and model parameters and translates them into a Java representation that is amenable for analysis using the Symbolic PathFinder symbolic execution tool. Notably, NEUROSPF encodes specialized peer classes for parsing the model's parameters, thereby enabling efficient analysis. With NEUROSPF the user has the flexibility to specify either the inputs or the network internal parameters as symbolic, promoting the application of program analysis and testing approaches from software engineering to the field of machine learning. For instance, NEUROSPF can be used for coverage-based testing and test generation, finding adversarial examples and also constraint-based repair of neural networks, thus improving the reliability of neural networks and of the applications that use them. Video URL: https://youtu.be/seal8fG78LI
ISBN:1665412194
9781665412193
DOI:10.1109/ICSE-Companion52605.2021.00027