Visual Probing: Cognitive Framework for Explaining Self-Supervised Image Representations
arXiv:2106.11054 · doi:10.1109/ACCESS.2023.3242982
Abstract
Recently introduced self-supervised methods for image representation learning provide on par or superior results to their fully supervised competitors, yet the corresponding efforts to explain the self-supervised approaches lag behind. Motivated by this observation, we introduce a novel visual probing framework for explaining the self-supervised models by leveraging probing tasks employed previously in natural language processing. The probing tasks require knowledge about semantic relationships between image parts. Hence, we propose a systematic approach to obtain analogs of natural language in vision, such as visual words, context, and taxonomy. Our proposal is grounded in Marr's computational theory of vision and concerns features like textures, shapes, and lines. We show the effectiveness and applicability of those analogs in the context of explaining self-supervised representations. Our key findings emphasize that relations between language and vision can serve as an effective yet intuitive tool for discovering how machine learning models work, independently of data modality. Our work opens a plethora of research pathways towards more explainable and transparent AI.
arXiv commentSubmitted to IEEE Access
References in corpus (10)
- Bootstrap your own latent: A new approach to self-supervised Learning
- Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
- Hierarchical Text-Conditional Image Generation with CLIP Latents
- Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
- Unsupervised Representation Learning by Predicting Image Rotations
- This Looks Like That: Deep Learning for Interpretable Image Recognition
- Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty
- ProtoPShare: Prototype Sharing for Interpretable Image Classification and Similarity Discovery
- When Explanations Lie: Why Many Modified BP Attributions Fail
- On the surprising similarities between supervised and self-supervised models