Automatic Concept Embedding Model (ACEM): No train-time concepts, No issue!
This addresses the problem of high annotation costs for interpretable models in safety-critical domains, but it appears incremental as it builds directly on existing CEMs.
The paper tackles the limitation of Concept Embedding Models (CEMs) requiring expensive concept annotations for all training data by proposing Automatic Concept Embedding Models (ACEMs) that learn these annotations automatically, though no concrete numbers are provided.
Interpretability and explainability of neural networks is continuously increasing in importance, especially within safety-critical domains and to provide the social right to explanation. Concept based explanations align well with how humans reason, proving to be a good way to explain models. Concept Embedding Models (CEMs) are one such concept based explanation architectures. These have shown to overcome the trade-off between explainability and performance. However, they have a key limitation -- they require concept annotations for all their training data. For large datasets, this can be expensive and infeasible. Motivated by this, we propose Automatic Concept Embedding Models (ACEMs), which learn the concept annotations automatically.