Welcome to the InfoSci Platform

Reference Hub

This research has been cited in:

Conference
Layer-Wise Analysis of Neuron Activation Values for Performance Verification of Artificial Neural Network Classifiers2022 IEEE International Conference on Assured Autonomy (ICAA)10.1109/ICAA52185.2022.00016
Conference
Uncertainty Quantification for Machine Learning Output Assurance using Anomaly-based Dataset Dissimilarity Measures2023 IEEE International Conference On Artificial Intelligence Testing (AITest)10.1109/AITest58265.2023.00013

An Integrated Process for Verifying Deep Learning Classifiers Using Dataset Dissimilarity Measures

Darryl Hond, Hamid Asgari, Daniel Jeffery, Mike Newman

Source Title: International Journal of Artificial Intelligence and Machine Learning (IJAIML)11(2)

ISSN: 2642-1577|EISSN: 2642-1585|EISBN13: 9781799864110|DOI: 10.4018/IJAIML.289536

Cite Article Cite Article

MLA

Hond, Darryl, et al. "An Integrated Process for Verifying Deep Learning Classifiers Using Dataset Dissimilarity Measures." IJAIML vol.11, no.2 2021: pp.1-21. http://doi.org/10.4018/IJAIML.289536

APA

Hond, D., Asgari, H., Jeffery, D., & Newman, M. (2021). An Integrated Process for Verifying Deep Learning Classifiers Using Dataset Dissimilarity Measures. International Journal of Artificial Intelligence and Machine Learning (IJAIML), 11(2), 1-21. http://doi.org/10.4018/IJAIML.289536

Chicago

Hond, Darryl, et al. "An Integrated Process for Verifying Deep Learning Classifiers Using Dataset Dissimilarity Measures," International Journal of Artificial Intelligence and Machine Learning (IJAIML) 11, no.2: 1-21. http://doi.org/10.4018/IJAIML.289536

Export Reference

Favorite Full-Issue Download

View Full Text HTML

View Full Text PDF

Abstract

The specification and verification of algorithms is vital for safety-critical autonomous systems which incorporate deep learning elements. We propose an integrated process for verifying artificial neural network (ANN) classifiers. This process consists of an off-line verification and an on-line performance prediction phase. The process is intended to verify ANN classifier generalisation performance, and to this end makes use of dataset dissimilarity measures. We introduce a novel measure for quantifying the dissimilarity between the dataset used to train a classification algorithm, and the test dataset used to evaluate and verify classifier performance. A system-level requirement could specify the permitted form of the functional relationship between classifier performance and a dissimilarity measure; such a requirement could be verified by dynamic testing. Experimental results, obtained using publicly available datasets, suggest that the measures have relevance to real-world practice for both quantifying dataset dissimilarity, and specifying and verifying classifier performance.