Label-free Concept Bottleneck Models

Abstract

Concept bottleneck model (CBM) are a popular way of creating more interpretable neural network by having hidden layer neurons correspond to human-understandable concepts. However, existing CBMs and their variants have two crucial limitations: first, the need to collect labeled data for each of the predefined concepts, which is time consuming and labor intensive; second, the accuracy of a CBM is often significantly lower than that of a standard neural network, especially on more complex datasets. This poor performance creates a barrier for adoption in practical real world applications. Motivated by these challenges, we propose Label-free CBM which is a framework to transform any neural network into an interpretable CBM without labeled concept data, while retaining a high accuracy. Our Label-free CBM has many advantages, it is: scalable - we present the first CBM scaled to ImageNet, efficient - creating a CBM takes only a few hours even for very large datasets, and automated - training it for a new dataset requires minimal human effort.