Towards Shape Biased Unsupervised Representation Learning for Domain Generalization

Abstract

It is known that, without awareness of the process, our brain appears to focus on the general shape of objects rather than superficial statistics of context. On the other hand, learning autonomously allows discovering invariant regularities which help generalization. In this work, we propose a learning framework to improve the shape bias property of self-supervised methods. Our method learns semantic and shape biased representations by integrating domain diversification and jigsaw puzzles. The first module enables the model to create a dynamic environment across arbitrary domains and provides a domain exploration vs. exploitation trade-off, while the second module allows the model to explore this environment autonomously. This universal framework does not require prior knowledge of the domain of interest. Extensive experiments are conducted on several domain generalization datasets, namely, PACS, Office-Home, VLCS, and Digits. We show that our framework outperforms state-of-the-art domain generalization methods by a large margin.

arXiv commentUnder review

References in corpus (11)

• Learning Transferable Features with Deep Adaptation Networks
• Deep Visual Domain Adaptation: A Survey
• Conditional Adversarial Domain Adaptation
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks
• CyCADA: Cycle-Consistent Adversarial Domain Adaptation
• Asymmetric Tri-training for Unsupervised Domain Adaptation
• Generalisation in humans and deep neural networks
• Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey
• Learning to Generalize: Meta-Learning for Domain Generalization
• Self-ensembling for visual domain adaptation
• Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation

Cited by in corpus (10)

• In Search of Lost Domain Generalization
• Domain Generalization using Causal Matching
• Understanding the Failure Modes of Out-of-Distribution Generalization
• Auxiliary Task Reweighting for Minimum-data Learning
• Model-Based Domain Generalization
• The Connection between Out-of-Distribution Generalization and Privacy of ML Models
• Robust Representation Learning via Perceptual Similarity Metrics
• Explainability-aided Domain Generalization for Image Classification
• Combining Diverse Feature Priors
• Discriminative Domain-Invariant Adversarial Network for Deep Domain Generalization