Interpretable Image Classification with Adaptive Prototype-based Vision Transformers
This work addresses the need for more interpretable and accurate image classification models, particularly for domains requiring transparency, but it appears incremental as it builds on existing prototype-based methods with enhancements.
The paper tackles the problem of interpretable image classification by introducing ProtoViT, which combines Vision Transformers with adaptive, deformable prototypes to improve accuracy and provide clear explanations. The result shows that the model generally achieves higher performance than existing prototype-based models, though no specific numbers are provided.
We present ProtoViT, a method for interpretable image classification combining deep learning and case-based reasoning. This method classifies an image by comparing it to a set of learned prototypes, providing explanations of the form ``this looks like that.'' In our model, a prototype consists of \textit{parts}, which can deform over irregular geometries to create a better comparison between images. Unlike existing models that rely on Convolutional Neural Network (CNN) backbones and spatially rigid prototypes, our model integrates Vision Transformer (ViT) backbones into prototype based models, while offering spatially deformed prototypes that not only accommodate geometric variations of objects but also provide coherent and clear prototypical feature representations with an adaptive number of prototypical parts. Our experiments show that our model can generally achieve higher performance than the existing prototype based models. Our comprehensive analyses ensure that the prototypes are consistent and the interpretations are faithful.