@inproceedings{6f0289a33f3e4e09a7cb7e6d606fd2f5,
title = "NEUROSPF: A Tool for the Symbolic Analysis of Neural Networks",
abstract = "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/seal8fG78L.",
keywords = "Neural Networks, Symbolic Execution, Symbolic PathFinder",
author = "Muhammad Usman and Yannic Noller and Pasareanu, {Corina S.} and Youcheng Sun and Divya Gopinath",
note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 43rd IEEE/ACM International Conference on Software Engineering: Companion, ICSE-Companion 2021 ; Conference date: 25-05-2021 Through 28-05-2021",
year = "2021",
month = may,
doi = "10.1109/ICSE-Companion52605.2021.00027",
language = "English",
series = "Proceedings - International Conference on Software Engineering",
publisher = "IEEE Computer Society ",
pages = "25--28",
booktitle = "Proceedings - 2021 IEEE/ACM 43rd International Conference on Software Engineering",
address = "United States",
}