Generating Post-Hoc Rationales of Deep Visual Classification Decisions

Abstract

Clearly explaining a rationale for a classification decision to an end-user can be as important as the decision itself. Existing approaches for deep visual recognition are generally opaque and do not output any justification text; contemporary vision-language models can describe image content but fail to take into account class-discriminative image aspects which justify visual predictions. Our model focuses on the discriminating properties of the visible object, jointly predicts a class label, and explains why the predicted label is appropriate for the image. A sampling and reinforcement learning based loss function learns to generate sentences that realize a global sentence property, such as class specificity. Our results on a fine-grained bird species classification dataset show that this model is able to generate explanations which are not only consistent with an image but also more discriminative than descriptions produced by existing captioning methods. In this work, we emphasize the importance of producing an explanation for an observed action, which could be applied to a black-box decision agent, akin to what one human produces when asked to explain the actions of a second human.