Episodic Training for Domain Generalization

Abstract

Domain generalization (DG) is the challenging and topical problem of learning models that generalize to novel testing domains with different statistics than a set of known training domains. The simple approach of aggregating data from all source domains and training a single deep neural network end-to-end on all the data provides a surprisingly strong baseline that surpasses many prior published methods. In this paper, we build on this strong baseline by designing an episodic training procedure that trains a single deep network in a way that exposes it to the domain shift that characterises a novel domain at runtime. Specifically, we decompose a deep network into feature extractor and classifier components, and then train each component by simulating it interacting with a partner who is badly tuned for the current domain. This makes both components more robust, ultimately leading to our networks producing state-of-the-art performance on three DG benchmarks. Furthermore, we consider the pervasive workflow of using an ImageNet trained CNN as a fixed feature extractor for downstream recognition tasks. Using the Visual Decathlon benchmark, we demonstrate that our episodic-DG training improves the performance of such a general-purpose feature extractor by explicitly training a feature for robustness to novel problems. This shows that DG training can benefit standard practice in computer vision.

arXiv commentICCV'19 CR version and fix Table 5. Code is now available at https://github.com/HAHA-DL/Episodic-DG

References in corpus (13)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Learning Transferable Features with Deep Adaptation Networks
• Deep Domain Confusion: Maximizing for Domain Invariance
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Domain Separation Networks
• Domain Generalization via Invariant Feature Representation
• Learning multiple visual domains with residual adapters
• Generalizing Across Domains via Cross-Gradient Training
• Universal representations:The missing link between faces, text, planktons, and cat breeds
• Feature-Critic Networks for Heterogeneous Domain Generalization

Cited by in corpus (13)

• Generalizing to unseen domains via distribution matching
• In Search of Lost Domain Generalization
• Test-Time Training with Self-Supervision for Generalization under Distribution Shifts
• Domain Generalization with MixStyle
• MixStyle Neural Networks for Domain Generalization and Adaptation
• Variational Disentanglement for Domain Generalization
• Semi-Supervised Domain Generalizable Person Re-Identification
• Deep Domain-Adversarial Image Generation for Domain Generalisation
• Latent Domain Learning with Dynamic Residual Adapters
• Learning to Diversify for Single Domain Generalization
• EEG-Based Driver Drowsiness Estimation Using Feature Weighted Episodic Training
• Neuron Coverage-Guided Domain Generalization
• Exploring Data Aggregation and Transformations to Generalize across Visual Domains