References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: PRIME: Prioritizing Interpretability in Failure Mode Extraction
(Submitted on 29 Sep 2023 (v1), last revised 14 Mar 2024 (this version, v2))
Abstract: In this work, we study the challenge of providing human-understandable descriptions for failure modes in trained image classification models. Existing works address this problem by first identifying clusters (or directions) of incorrectly classified samples in a latent space and then aiming to provide human-understandable text descriptions for them. We observe that in some cases, describing text does not match well with identified failure modes, partially owing to the fact that shared interpretable attributes of failure modes may not be captured using clustering in the feature space. To improve on these shortcomings, we propose a novel approach that prioritizes interpretability in this problem: we start by obtaining human-understandable concepts (tags) of images in the dataset and then analyze the model's behavior based on the presence or absence of combinations of these tags. Our method also ensures that the tags describing a failure mode form a minimal set, avoiding redundant and noisy descriptions. Through several experiments on different datasets, we show that our method successfully identifies failure modes and generates high-quality text descriptions associated with them. These results highlight the importance of prioritizing interpretability in understanding model failures.
Submission history
From: Keivan Rezaei [view email][v1] Fri, 29 Sep 2023 22:00:12 GMT (41942kb,D)
[v2] Thu, 14 Mar 2024 04:29:54 GMT (44267kb,D)
Link back to: arXiv, form interface, contact.