Timezone: »

 
Provable Concept Learning for Interpretable Predictions Using Variational Autoencoders
Armeen Taeb · Nicolò Ruggeri · Carina Schnuck · Fanny Yang

@
in AI for Science »
Event URL: https://openreview.net/forum?id=0sBvXqyvgUT »

In safety-critical applications, practitioners are reluctant to trust neural networks when no interpretable explanations are available. Many attempts to provide such explanations revolve around pixel-level attributions or use previously known concepts. In this paper we aim to provide explanations by provably identifying \emph{high-level, previously unknown concepts}. To this end, we propose a probabilistic modeling framework to derive (C)oncept (L)earning and (P)rediction (CLAP) - a VAE-based classifier that uses visually interpretable concepts as linear predictors. Assuming that the data generating mechanism involves interpretable concepts, we prove that our method is able to identify them while attaining optimal classification accuracy. We use synthetic experiments for validation, and also show that on the ChestXRay dataset, CLAP effectively discovers interpretable factors for classifying diseases.

Author Information

Armeen Taeb (Swiss Federal Institute of Technology)
Nicolò Ruggeri (Max-Planck Institute / ETH)
Carina Schnuck (ETHZ - ETH Zurich)
Fanny Yang (ETH Zurich)

More from the Same Authors