Navigating Neural Space: Revisiting Concept Activation Vectors to Overcome Directional Divergence

With a growing interest in understanding neural network prediction strategies, Concept Activation Vectors (CAVs) have emerged as a popular tool for modeling human-understandable concepts in the latent space. Commonly, CAVs are computed by leveraging linear classifiers optimizing the separability of latent representations of samples with and without a given concept. However, in this paper we show that such a separability-oriented computation leads to solutions, which may diverge from the actual goal of precisely modeling the concept direction. This discrepancy can …

classifiers concept concepts cs.ai cs.cv cs.lg divergence human linear modeling network neural network popular prediction samples space strategies tool understanding vectors