Quantifying the Accuracy-Interpretability Trade-Off in Concept-Based Sidechannel Models

Open in new window