A Framework for Causal Concept-based Model Explanations

TL;DR

This paper introduces a causal concept-based framework for post-hoc explanations of AI models, emphasizing understandability and fidelity through concept interventions and causal interpretation, demonstrated on CelebA classifiers.

Contribution

It proposes a novel framework for causal, concept-based explanations that balances interpretability and faithfulness, with a proof-of-concept implementation.

Findings

Framework enables local and global explanations using concept interventions.

Explainability is achieved through a clear, causal concept vocabulary.

Fidelity depends on aligning explanation interpretation with model context.

Abstract

This work presents a conceptual framework for causal concept-based post-hoc Explainable Artificial Intelligence (XAI), based on the requirements that explanations for non-interpretable models should be understandable as well as faithful to the model being explained. Local and global explanations are generated by calculating the probability of sufficiency of concept interventions. Example explanations are presented, generated with a proof-of-concept model made to explain classifiers trained on the CelebA dataset. Understandability is demonstrated through a clear concept-based vocabulary, subject to an implicit causal interpretation. Fidelity is addressed by highlighting important framework assumptions, stressing that the context of explanation interpretation must align with the context of explanation generation.