Towards Robust Metrics for Concept Representation Evaluation


  • Mateo Espinosa Zarlenga University of Cambridge
  • Pietro Barbiero University of Cambridge
  • Zohreh Shams University of Cambridge Babylon Health
  • Dmitry Kazhdan University of Cambridge
  • Umang Bhatt University of Cambridge The Alan Turing Institute
  • Adrian Weller University of Cambridge The Alan Turing Institute
  • Mateja Jamnik University of Cambridge



PEAI: Interpretability and Explainability, PEAI: Safety, Robustness & Trustworthiness, ML: Representation Learning, ML: Deep Generative Models & Autoencoders


Recent work on interpretability has focused on concept-based explanations, where deep learning models are explained in terms of high-level units of information, referred to as concepts. Concept learning models, however, have been shown to be prone to encoding impurities in their representations, failing to fully capture meaningful features of their inputs. While concept learning lacks metrics to measure such phenomena, the field of disentanglement learning has explored the related notion of underlying factors of variation in the data, with plenty of metrics to measure the purity of such factors. In this paper, we show that such metrics are not appropriate for concept learning and propose novel metrics for evaluating the purity of concept representations in both approaches. We show the advantage of these metrics over existing ones and demonstrate their utility in evaluating the robustness of concept representations and interventions performed on them. In addition, we show their utility for benchmarking state-of-the-art methods from both families and find that, contrary to common assumptions, supervision alone may not be sufficient for pure concept representations.




How to Cite

Espinosa Zarlenga, M., Barbiero, P., Shams, Z., Kazhdan, D., Bhatt, U., Weller, A., & Jamnik, M. (2023). Towards Robust Metrics for Concept Representation Evaluation. Proceedings of the AAAI Conference on Artificial Intelligence, 37(10), 11791-11799.



AAAI Technical Track on Philosophy and Ethics of AI