Learnable Visual Words for Interpreting Image Recognition Models

Wenxiao Xiao, Zhengming Ding, Hongfu Liu

Abstract: To interpret deep models' predictions, attention-based visual cues are widely used in addressing *why* deep models make such predictions. Beyond that, the current research community becomes more interested in reasoning *how* deep models make predictions, where some prototype-based methods employ interpretable representations with their corresponding visual cues to reveal the black-box mechanism of deep model behaviors. However, these pioneering attempts only either learn the category-specific prototypes and deteriorate with their generalization ability deterioration or demonstrate several illustrative examples without a quantitative evaluation of visual-based interpretability narrowing their practical usages. In this paper, we revisit the concept of visual words and propose the Learnable Visual Words (LVW) to interpret the model prediction behaviors with two novel modules: semantic visual words learning and dual fidelity preservation. The semantic visual words learning relaxes the category-specific constraint, enabling the generic visual words shared across multiple categories. Beyond employing the visual words for prediction to align visual words with the base model, our dual fidelity preservation also includes the attention-guided semantic alignment that encourages the learned visual words to focus on the same conceptual regions for prediction. Experiments on six visual benchmarks demonstrate the superior effectiveness of our proposed LVW in both accuracy and interpretation fidelity over the state-of-the-art methods. Moreover, we elaborate on various in-depth analyses to further explore the learned visual words and the generalizability of our method for unseen categories.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning