Interpretable Generative Models through Post-hoc Concept Bottlenecks
This work addresses the problem of creating interpretable generative models for researchers and practitioners, offering a more efficient and scalable solution compared to existing approaches.
The paper tackled the inefficiency and scalability issues in building interpretable generative models by introducing two low-cost post-hoc methods, concept-bottleneck autoencoder and concept controller, which achieved an average improvement of ~25% over prior work and were 4-15x faster to train.
Concept bottleneck models (CBM) aim to produce inherently interpretable models that rely on human-understandable concepts for their predictions. However, existing approaches to design interpretable generative models based on CBMs are not yet efficient and scalable, as they require expensive generative model training from scratch as well as real images with labor-intensive concept supervision. To address these challenges, we present two novel and low-cost methods to build interpretable generative models through post-hoc techniques and we name our approaches: concept-bottleneck autoencoder (CB-AE) and concept controller (CC). Our proposed approaches enable efficient and scalable training without the need of real data and require only minimal to no concept supervision. Additionally, our methods generalize across modern generative model families including generative adversarial networks and diffusion models. We demonstrate the superior interpretability and steerability of our methods on numerous standard datasets like CelebA, CelebA-HQ, and CUB with large improvements (average ~25%) over the prior work, while being 4-15x faster to train. Finally, a large-scale user study is performed to validate the interpretability and steerability of our methods.