Anchors vs Attention: Comparing XAI on a Real-Life Use Case

Gaëlle Jouis (1, 2, 3) , Harold Mouchère (1, 2) , Fabien Picarougne (1, 4) , Alexandre Hardouin (3)

1 LS2N - Laboratoire des Sciences du Numérique de Nantes
2 LS2N - équipe IPI - Image Perception Interaction
3 Pôle emploi (France)
4 LS2N - équipe DUKe - Data User Knowledge

Recent advances in eXplainable Artificial Intelligence (XAI) led to many different methods in order to improve explainability of deep learning algorithms. With many options at hand, and maybe the need to adapt existing ones to new problems, one may find in a struggle to choose the right method to generate explanations. This paper presents an objective approach to compare two different existing XAI methods. These methods are applied to a use case from literature and to a real use case of a French administration.

Mots clés: Machine learning; Deep learning; Explainability; Attention networks

Domaines: Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]; Réseau de neurones [cs.NE]

Gaëlle Jouis, Harold Mouchère, Fabien Picarougne, Alexandre Hardouin. Anchors vs Attention: Comparing XAI on a Real-Life Use Case. ICPR 2021: Pattern Recognition. ICPR International Workshops and Challenges, Jan 2021, Virtual, France. pp.219-227, ⟨10.1007/978-3-030-68796-0_16⟩. ⟨hal-03210595⟩