Improving Intervention Efficacy via Concept Realignment in Concept Bottleneck Models

arXiv:2405.01531v1 Announce Type: new
Abstract: Concept Bottleneck Models (CBMs) ground image classification on human-understandable concepts to allow for interpretable model decisions. Crucially, the CBM design inherently allows for human interventions, in which expert users are given the ability to modify potentially misaligned concept choices to influence the decision behavior of the model in an interpretable fashion. However, existing approaches often require numerous human interventions per image to achieve strong performances, posing practical challenges in scenarios where obtaining human feedback is …

abstract arxiv behavior classification concept concepts cs.ai cs.cv cs.lg decision decisions design expert human image improving influence type via