Generalizing to Unseen Domains via Adversarial Data Augmentation

2018/05/30 by Riccardo Volpi, Volpi, Riccardo, Hongseok Namkoong +9 · 43 citations
Computer Science · Medicine · #Adversarial Robustness in Machine Learning #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.1805.12018

openalex publication_date 2018/05/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We are concerned with learning models that generalize well to different unseen domains. We consider a worst-case formulation over data distributions that are near the source domain in the feature space. Only using training data from a single source distribution, we propose an iterative procedure that augments the dataset with examples from a fictitious target domain that is "hard" under the current model. We show that our iterative scheme is an adaptive data augmentation method where we append adversarial examples at each iteration. For softmax losses, we show that our method is a data-dependent regularization scheme that behaves differently from classical regularizers that regularize towards zero (e.g., ridge or lasso). On digit recognition and semantic segmentation tasks, our method learns models improve performance across a range of a priori unknown target domains.

Citations

• Gradient-based learning applied to document recognition
• Backpropagation Applied to Handwritten Zip Code Recognition
• Certifying Some Distributional Robustness with Principled Adversarial\n Training
• Adversarial Discriminative Domain Adaptation
• Fully Convolutional Networks for Semantic Segmentation

Cited by

• Fake News Classification in Urdu: A Domain Adaptation Approach for a Low-Resource Language
• Federated Domain Generalization with Latent Space Inversion
• HydroDCM: Hydrological Domain-Conditioned Modulation for Cross-Reservoir Inflow Prediction
• A Probabilistic Framework for Temporal Distribution Generalization in Industry-Scale Recommender Systems
• Modality-Balanced Collaborative Distillation for Multi-Modal Domain Generalization
• A Theory-Inspired Framework for Few-Shot Cross-Modal Sketch Person Re-Identification
• ODP-Bench: Benchmarking Out-of-Distribution Performance Prediction
• C-LEAD: Contrastive Learning for Enhanced Adversarial Defense
• Mitigating Spurious Correlation via Distributionally Robust Learning with Hierarchical Ambiguity Sets
• AdvBlur: Adversarial Blur for Robust Diabetic Retinopathy Classification and Cross-Domain Generalization
• Humanoid-inspired Causal Representation Learning for Domain Generalization
• Environment Inference for Learning Generalizable Dynamical System
• MEASURE: Multi-scale Minimal Sufficient Representation Learning for Domain Generalization in Sleep Staging
• Variational Disentanglement for Domain Generalization
• Analytical Survey of Learning with Low-Resource Data: From Analysis to Investigation
• Robustness through Data Augmentation Loss Consistency
• Diffusion-Based Action Recognition Generalizes to Untrained Domains
• Unsupervised Domain Adaptation in the Absence of Source Data
• ConstStyle: Robust Domain Generalization with Unified Style Transformation
• Generalizing Nucleus Recognition Model in Multi-source Images via Pruning
• Towards Out-Of-Distribution Generalization: A Survey
• Target-Oriented Single Domain Generalization
• MorphGen: Morphology-Guided Representation Learning for Robust Single-Domain Generalization in Histopathological Cancer Classification
• Improving OCR using internal document redundancy
• Class Unbiasing for Generalization in Medical Diagnosis
• Set Pivot Learning: Redefining Generalized Segmentation with Vision Foundation Models
• On the effectiveness of adversarial training against common corruptions
• Adversarial Training Improves Generalization Under Distribution Shifts in Bioacoustics
• Domain generalization of 3D semantic segmentation in autonomous driving
• Contrastive ACE: Domain Generalization Through Alignment of Causal Mechanisms
• Confidence Regions in Wasserstein Distributionally Robust Estimation
• Leveraging Depth and Language for Open-Vocabulary Domain-Generalized Semantic Segmentation
• Text-Driven Causal Representation Learning for Source-Free Domain Generalization
• An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation
• Rule Learning for Knowledge Graph Reasoning under Agnostic Distribution Shift
• Adversarial Data Augmentation for Single Domain Generalization via Lyapunov Exponent-Guided Optimization
• Adapting to Unseen Vendor Domains for MRI Lesion Segmentation
• Residual Reweighted Conformal Prediction for Graph Neural Networks
• Dealing with the Evil Twins: Improving Random Augmentation by Addressing Catastrophic Forgetting of Diverse Augmentations
• Generalizing to New Dynamical Systems via Frequency Domain Adaptation
• Domain Generalization for Person Re-identification: A Survey Towards Domain-Agnostic Person Matching
• Human Imperceptible Attacks and Applications to Improve Fairness
• Learning Less Generalizable Patterns with an Asymmetrically Trained Double Classifier for Better Test-Time Adaptation

Related

• Learning to Learn Single Domain Generalization
• Domain Generalization by Rejecting Extreme Augmentations
• When Neural Networks Fail to Generalize? A Model Sensitivity Perspective
• Improving Adversarial Robustness via Unlabeled Out-of-Domain Data
• Adversarial Dropout Regularization