Skip to main content

Research Repository

Advanced Search

Agree to Disagree: When Deep Learning Models With Identical Architectures Produce Distinct Explanations

Watson, M.; Awwad Shiekh Hasan, B.; Al Moubayed, N.

Agree to Disagree: When Deep Learning Models With Identical Architectures Produce Distinct Explanations Thumbnail


Profile Image

Matthew Watson
Postdoctoral Research Associate


Deep Learning of neural networks has progressively become more prominent in healthcare with models reaching, or even surpassing, expert accuracy levels. However, these success stories are tainted by concerning reports on the lack of model transparency and bias against some medical conditions or patients’ sub-groups. Explainable methods are considered the gateway to alleviate many of these concerns. In this study we demonstrate that the generated explanations are volatile to changes in model training that are perpendicular to the classification task and model structure. This raises further questions about trust in deep learning models for healthcare. Mainly, whether the models capture underlying causal links in the data or just rely on spurious correlations that are made visible via explanation methods. We demonstrate that the output of explainability methods on deep neural networks can vary significantly by changes of hyper-parameters, such as the random seed or how the training set is shuffled. We introduce a measure of explanation consistency which we use to highlight the identified problems on the MIMIC-CXR dataset. We find explanations of identical models but with different training setups have a low consistency: ≈ 33% on average. On the contrary, kernel methods are robust against any orthogonal changes, with explanation consistency at 94%. We conclude that current trends in model explanation are not sufficient to mitigate the risks of deploying models in real life healthcare applications

Presentation Conference Type Conference Paper (Published)
Conference Name Proc. Winter Conference on Applications of Computer Vision
Start Date Jan 3, 2022
End Date Jan 8, 2022
Acceptance Date Oct 4, 2021
Online Publication Date Feb 15, 2022
Publication Date 2022
Deposit Date Oct 27, 2021
Publicly Available Date Jan 9, 2022
Publisher Institute of Electrical and Electronics Engineers
Public URL


Accepted Conference Proceeding (1.6 Mb)

Copyright Statement
© 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

You might also like

Downloadable Citations