Calculating the Credibility of Test Samples at Inference by a Layer-wise Activation Cluster Analysis of Convolutional Neural Networks

Daniel Lehmann, Marc Ebner

2022

Abstract

A convolutional neural network model is able to achieve high classification performance on test samples at inference, as long as those samples are drawn from the same distribution as the samples used for model training. However, if a test sample is drawn from a different distribution, the performance of the model decreases drastically. Such a sample is typically referred to as an out-of-distribution (OOD) sample. Papernot and McDaniel (2018) propose a method, called Deep k-Nearest Neighbors (DkNN), to detect OOD samples by a credibility score. However, DkNN are slow at inference as they are based on a kNN search. To address this problem, we propose a detection method that uses clustering instead of a kNN search. We conducted experiments with different types of OOD samples for models trained on either MNIST, SVHN, or CIFAR10. Our experiments show that our method is significantly faster than DkNN, while achieving similar performance.

Paper Citation

in Harvard Style

Lehmann D. and Ebner M. (2022). Calculating the Credibility of Test Samples at Inference by a Layer-wise Activation Cluster Analysis of Convolutional Neural Networks. In Proceedings of the 3rd International Conference on Deep Learning Theory and Applications - Volume 1: DeLTA, ISBN 978-989-758-584-5, pages 34-43. DOI: 10.5220/0011274000003277

in Bibtex Style

@conference{delta22,
author={Daniel Lehmann and Marc Ebner},
title={Calculating the Credibility of Test Samples at Inference by a Layer-wise Activation Cluster Analysis of Convolutional Neural Networks},
booktitle={Proceedings of the 3rd International Conference on Deep Learning Theory and Applications - Volume 1: DeLTA,},
year={2022},
pages={34-43},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011274000003277},
isbn={978-989-758-584-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Deep Learning Theory and Applications - Volume 1: DeLTA,
TI - Calculating the Credibility of Test Samples at Inference by a Layer-wise Activation Cluster Analysis of Convolutional Neural Networks
SN - 978-989-758-584-5
AU - Lehmann D.
AU - Ebner M.
PY - 2022
SP - 34
EP - 43
DO - 10.5220/0011274000003277