This actually looks like that: Proto-BagNets for local and global interpretability-by-design
This work addresses interpretability needs in high-stakes applications like medical diagnosis by offering a more faithful and clinically meaningful explanation method, though it appears incremental as it builds on existing prototype-based and bag-of-local feature approaches.
The authors tackled the problem of unreliable explanations in prototype-based networks by introducing Proto-BagNets, which combines bag-of-local features with prototype learning to provide meaningful and accurate explanations for image classification, achieving performance comparable to state-of-the-art models in drusen detection on retinal OCT data.
Interpretability is a key requirement for the use of machine learning models in high-stakes applications, including medical diagnosis. Explaining black-box models mostly relies on post-hoc methods that do not faithfully reflect the model's behavior. As a remedy, prototype-based networks have been proposed, but their interpretability is limited as they have been shown to provide coarse, unreliable, and imprecise explanations. In this work, we introduce Proto-BagNets, an interpretable-by-design prototype-based model that combines the advantages of bag-of-local feature models and prototype learning to provide meaningful, coherent, and relevant prototypical parts needed for accurate and interpretable image classification tasks. We evaluated the Proto-BagNet for drusen detection on publicly available retinal OCT data. The Proto-BagNet performed comparably to the state-of-the-art interpretable and non-interpretable models while providing faithful, accurate, and clinically meaningful local and global explanations. The code is available at https://github.com/kdjoumessi/Proto-BagNets.