Skip to yearly menu bar Skip to main content


Poster

Understanding Inter-Concept Relationships in Concept-Based Models

Naveen Raman · Mateo Espinosa Zarlenga · Mateja Jamnik

Hall C 4-9 #2413
[ ] [ Paper PDF ]
[ Poster
Wed 24 Jul 2:30 a.m. PDT — 4 a.m. PDT

Abstract:

Concept-based explainability methods provide insight into deep learning systems by constructing explanations using human-understandable concepts. While the literature on human reasoning demonstrates that we exploit relationships between concepts when solving tasks, it is unclear whether concept-based methods incorporate the rich structure of inter-concept relationships. We analyse the concept representations learnt by concept-based models to understand whether these models correctly capture inter-concept relationships. First, we empirically demonstrate that state-of-the-art concept-based models produce representations that lack stability and robustness, and such methods fail to capture inter-concept relationships. Then, we develop a novel algorithm which leverages inter-concept relationships to improve concept intervention accuracy, demonstrating how correctly capturing inter-concept relationships can improve downstream tasks.

Chat is not available.