Faithful Counterfactual Visual Explanations (FCVE)
Article
Khan, B., Tariq, S., Zia, T., Ahsan, M. and Windridge, D. 2024. Faithful Counterfactual Visual Explanations (FCVE). Knowledge-Based Systems. 294. https://doi.org/10.1016/j.knosys.2024.111668
Type | Article |
---|---|
Title | Faithful Counterfactual Visual Explanations (FCVE) |
Authors | Khan, B., Tariq, S., Zia, T., Ahsan, M. and Windridge, D. |
Abstract | Deep learning models in computer vision have made remarkable progress, but their lack of transparency and interpretability remains a challenge. The development of explainable AI can enhance the understanding and performance of these models. However, existing techniques often struggle to provide convincing explanations that non-experts easily understand, and they cannot accurately identify models’ intrinsic decision-making processes. To address these challenges, we propose to develop a counterfactual explanation (CE) model that balances plausibility and faithfulness. This model generates easy-to-understand visual explanations by making minimum changes necessary in images without altering the pixel data. Instead, the proposed method identifies internal concepts and filters learned by models and leverages them to produce plausible counterfactual explanations. The provided explanations reflect the internal decision-making process of the model, thus ensuring faithfulness to the model. |
Keywords | Explainable AI; Visual explanation; Counterfactual |
Sustainable Development Goals | 9 Industry, innovation and infrastructure |
Middlesex University Theme | Creativity, Culture & Enterprise |
Research Group | Artificial Intelligence group |
Publisher | Elsevier |
Journal | Knowledge-Based Systems |
ISSN | 0950-7051 |
Electronic | 1872-7409 |
Publication dates | |
Online | 26 Mar 2024 |
21 Jun 2024 | |
Publication process dates | |
Submitted | 20 Nov 2023 |
Accepted | 18 Mar 2024 |
Deposited | 09 Apr 2024 |
Output status | Published |
Accepted author manuscript | License File Access Level Open |
Copyright Statement | © 2024. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ |
Digital Object Identifier (DOI) | https://doi.org/10.1016/j.knosys.2024.111668 |
Web of Science identifier | WOS:001229329700001 |
https://repository.mdx.ac.uk/item/116z33
Restricted files
Accepted author manuscript
34
total views1
total downloads1
views this month0
downloads this month