Towards Better Generalization and Interpretability in Unsupervised Concept-Based Models
This work addresses the need for more interpretable and generalizable AI models in image classification, though it appears incremental as it builds on existing concept-based approaches.
The paper tackles the problem of improving trustworthiness in deep neural networks by introducing an unsupervised concept-based model for image classification, which achieves better generalization than existing unsupervised models and nearly matches black-box performance while enhancing interpretability.
To increase the trustworthiness of deep neural networks, it is critical to improve the understanding of how they make decisions. This paper introduces a novel unsupervised concept-based model for image classification, named Learnable Concept-Based Model (LCBM) which models concepts as random variables within a Bernoulli latent space. Unlike traditional methods that either require extensive human supervision or suffer from limited scalability, our approach employs a reduced number of concepts without sacrificing performance. We demonstrate that LCBM surpasses existing unsupervised concept-based models in generalization capability and nearly matches the performance of black-box models. The proposed concept representation enhances information retention and aligns more closely with human understanding. A user study demonstrates the discovered concepts are also more intuitive for humans to interpret. Finally, despite the use of concept embeddings, we maintain model interpretability by means of a local linear combination of concepts.