Deep Dive into Deep Neural Networks with Flows

Adrien Halnaut, Romain Giot, Romain Bourqui, David Auber

2020

Abstract

Deep neural networks are becoming omnipresent in reason of their growing popularity in media and their daily use. However, their global complexity makes them hard to understand which emphasizes their black-box aspect and the lack of confidence given by their potential users. The use of tailored visual and interactive representations is one way to improve their explainability and trustworthy. Inspired by parallel coordinates and Sankey diagrams, this paper proposes a novel visual representation allowing tracing the progressive classification of a trained classification neural network by examining how each evaluation data is being processed by each network’s layer. It is thus possible to observe which data classes are quickly recognized, unstable, or lately recognized. Such information provides insights to the user about the model architecture’s pertinence and can guide on its improvement. The method has been validated on two classification neural networks inspired from the literature (LeNet5 and VGG16) using two public databases (MNIST and FashionMNIST).

Paper Citation

in Harvard Style

Halnaut A., Giot R., Bourqui R. and Auber D. (2020). Deep Dive into Deep Neural Networks with Flows. In Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 3: IVAPP; ISBN 978-989-758-402-2, SciTePress, pages 231-239. DOI: 10.5220/0008989702310239

in Bibtex Style

@conference{ivapp20,
author={Adrien Halnaut and Romain Giot and Romain Bourqui and David Auber},
title={Deep Dive into Deep Neural Networks with Flows},
booktitle={Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 3: IVAPP},
year={2020},
pages={231-239},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008989702310239},
isbn={978-989-758-402-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 3: IVAPP
TI - Deep Dive into Deep Neural Networks with Flows
SN - 978-989-758-402-2
AU - Halnaut A.
AU - Giot R.
AU - Bourqui R.
AU - Auber D.
PY - 2020
SP - 231
EP - 239
DO - 10.5220/0008989702310239
PB - SciTePress