EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainability and Evaluation of Vision Transformers: An In-Depth Experimental Study

Sédrick Stassin, Valentin Corduant, Sidi Ahmed Mahmoudi, Xavier Siebert

Electronics · 30.12.2023 · 13 (1) · 175

DOI: 10.3390/electronics13010175 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 30.12.2023. Verfasst von Sédrick Stassin und 3 weiteren Autor:innen. Die Arbeit wurde bisher 37-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Walloon Skywin pole of Belgium.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In the era of artificial intelligence (AI), the deployment of intelligent systems for autonomous decision making has surged across diverse fields. However, the widespread adoption of AI technology is hindered by the risks associated with ceding control to autonomous systems, particularly in critical domains. Explainable artificial intelligence (XAI) has emerged as a critical subdomain fostering human understanding and trust. It addresses the opacity of complex models such as vision transformers (ViTs), which have gained prominence lately. With the expanding landscape of XAI methods, selecting the most effective method remains an open question, due to the lack of a ground-truth label for explainability. To avoid subjective human judgment, numerous metrics have been developed, with each aiming to fulfill certain properties required for a valid explanation. This study conducts a detailed evaluation of various XAI methods applied to the ViT architecture, thereby exploring metrics criteria like faithfulness, coherence, robustness, and complexity. We especially study the metric convergence, correlation, discriminative power, and inference time of both XAI methods and metrics. Contrary to expectations, the metrics of each criterion reveal minimal convergence and correlation. This study not only challenges the conventional practice of metric-based ranking of XAI methods but also underscores the dependence of explanations on the experimental environment, thereby presenting crucial considerations for the future development and adoption of XAI methods in real-world applications.

Förderung

  • Walloon Skywin pole of Belgium

Zitieren

APA

Stassin, S., Corduant, V., Mahmoudi, S. A., & Siebert, X. (2023). Explainability and Evaluation of Vision Transformers: An In-Depth Experimental Study. Electronics, 13(1), 175. https://doi.org/10.3390/electronics13010175

Vancouver

Stassin S, Corduant V, Mahmoudi SA, Siebert X. Explainability and Evaluation of Vision Transformers: An In-Depth Experimental Study. Electronics. 2023;13(1):175. doi:10.3390/electronics13010175

BibTeX
@article{Stassin2023,
  author = {Stassin, Sédrick and Corduant, Valentin and Mahmoudi, Sidi Ahmed and Siebert, Xavier},
  title = {Explainability and Evaluation of Vision Transformers: An In-Depth Experimental Study},
  journal = {Electronics},
  year = {2023},
  volume = {13},
  number = {1},
  pages = {175},
  doi = {10.3390/electronics13010175}
}
Kostenlos starten →