The Query Channel: Information-Theoretic Limits of Masking-Based Explanations
Information theory reveals why popular model explanation tools fail to provide reliable insights
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 a...
ArXiv CS.AI
Source
Generated insight by OnAirToday AI
This research applies information theory to expose fundamental limits in how we interpret black-box AI models using popular techniques like SHAP and LIME. By framing explanations as a communication channel, the study proves that standard methods often fail to recover accurate feature importance when complexity exceeds the query budget.
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.