EQ-CBM: A Probabilistic Concept Bottleneck with Energy-based Models and Quantized Vectors

TL;DR

EQ-CBM introduces a probabilistic concept bottleneck framework utilizing energy-based models and quantized vectors, significantly improving interpretability, uncertainty estimation, and accuracy in deep neural networks.

Contribution

It presents a novel probabilistic concept encoding method with energy-based models and quantized vectors, enhancing interpretability and performance over existing CBMs.

Findings

Outperforms state-of-the-art in concept accuracy

Achieves higher task accuracy

Effectively captures uncertainties

Abstract

The demand for reliable AI systems has intensified the need for interpretable deep neural networks. Concept bottleneck models (CBMs) have gained attention as an effective approach by leveraging human-understandable concepts to enhance interpretability. However, existing CBMs face challenges due to deterministic concept encoding and reliance on inconsistent concepts, leading to inaccuracies. We propose EQ-CBM, a novel framework that enhances CBMs through probabilistic concept encoding using energy-based models (EBMs) with quantized concept activation vectors (qCAVs). EQ-CBM effectively captures uncertainties, thereby improving prediction reliability and accuracy. By employing qCAVs, our method selects homogeneous vectors during concept encoding, enabling more decisive task performance and facilitating higher levels of human intervention. Empirical results using benchmark datasets demonstrate that our approach outperforms the state-of-the-art in both concept and task accuracy.