Monitizer: Automating Design and Evaluation of Neural Network Monitors

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Azeem, Muqsit, Grobelna, Marta, Kanav, Sudeep, Kretinsky, Jan, Mohr, Stefanie, Rieder, Sabine
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916249957564416
author Azeem, Muqsit
Grobelna, Marta
Kanav, Sudeep
Kretinsky, Jan
Mohr, Stefanie
Rieder, Sabine
author_facet Azeem, Muqsit
Grobelna, Marta
Kanav, Sudeep
Kretinsky, Jan
Mohr, Stefanie
Rieder, Sabine
contents The behavior of neural networks (NNs) on previously unseen types of data (out-of-distribution or OOD) is typically unpredictable. This can be dangerous if the network's output is used for decision-making in a safety-critical system. Hence, detecting that an input is OOD is crucial for the safe application of the NN. Verification approaches do not scale to practical NNs, making runtime monitoring more appealing for practical use. While various monitors have been suggested recently, their optimization for a given problem, as well as comparison with each other and reproduction of results, remain challenging. We present a tool for users and developers of NN monitors. It allows for (i) application of various types of monitors from the literature to a given input NN, (ii) optimization of the monitor's hyperparameters, and (iii) experimental evaluation and comparison to other approaches. Besides, it facilitates the development of new monitoring approaches. We demonstrate the tool's usability on several use cases of different types of users as well as on a case study comparing different approaches from recent literature.
format Preprint
id arxiv_https___arxiv_org_abs_2405_10350
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Monitizer: Automating Design and Evaluation of Neural Network Monitors
Azeem, Muqsit
Grobelna, Marta
Kanav, Sudeep
Kretinsky, Jan
Mohr, Stefanie
Rieder, Sabine
Machine Learning
Artificial Intelligence
Software Engineering
The behavior of neural networks (NNs) on previously unseen types of data (out-of-distribution or OOD) is typically unpredictable. This can be dangerous if the network's output is used for decision-making in a safety-critical system. Hence, detecting that an input is OOD is crucial for the safe application of the NN. Verification approaches do not scale to practical NNs, making runtime monitoring more appealing for practical use. While various monitors have been suggested recently, their optimization for a given problem, as well as comparison with each other and reproduction of results, remain challenging. We present a tool for users and developers of NN monitors. It allows for (i) application of various types of monitors from the literature to a given input NN, (ii) optimization of the monitor's hyperparameters, and (iii) experimental evaluation and comparison to other approaches. Besides, it facilitates the development of new monitoring approaches. We demonstrate the tool's usability on several use cases of different types of users as well as on a case study comparing different approaches from recent literature.
title Monitizer: Automating Design and Evaluation of Neural Network Monitors
topic Machine Learning
Artificial Intelligence
Software Engineering
url https://arxiv.org/abs/2405.10350