arXiv:2604.16689v2 Announce Type: replace Abstract: Masking-based post-hoc explanation methods, such as KernelSHAP and LIME, estimate local feature importance by querying a black-box model under randomized perturbations. This paper formulates this procedure as communication over a query channel, where the latent explanation acts as a message and each masked evaluation is a channel use. Within this framework, the complexity of the explanation is captured by the entropy of the hypothesis class, while the query interface supplies information at a rate determined by an identification capacity per
The proliferation of black-box AI models necessitates robust explainability methods, making new theoretical foundations for these methods highly relevant.
A deeper information-theoretic understanding of AI explanation methods can lead to more reliable, efficient, and trustworthy AI systems, crucial for sensitive applications.
This research provides a novel framework for evaluating masking-based explanation techniques, potentially enabling the development of more principled and interpretable AI systems.
- · AI researchers
- · AI developers
- · Regulatory bodies
- · Developers relying on ad-hoc explainability methods
Improved theoretical understanding of AI explainability methods.
Development of more robust and auditable AI systems, especially in critical sectors.
Increased trust in AI applications across industries, accelerating adoption where explainability is a bottleneck.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI