Adversarial Bayesian Augmentation for Single-Source Domain Generalization

2023-07-18 · ICCV 2023 1 · Sheng Cheng, Tejas Gokhale, Yezhou Yang

Generalizing to unseen image domains is a challenging problem primarily due to the lack of diverse training data, inaccessible target data, and the large domain shift that may exist in many real-world settings. As such data augmentation is a critical component of domain generalization methods that seek to address this problem. We present Adversarial Bayesian Augmentation (ABA), a novel algorithm that learns to generate image augmentations in the challenging single-source domain generalization setting. ABA draws on the strengths of adversarial learning and Bayesian neural networks to guide the generation of diverse data augmentations -- these synthesized image domains aid the classifier in generalizing to unseen domains. We demonstrate the strength of ABA on several types of domain shift including style shift, subpopulation shift, and shift in the medical imaging setting. ABA outperforms all previous state-of-the-art methods, including pre-specified augmentations, pixel-based and convolutional-based augmentations.

📄 PDF Abstract BibTeX arXiv:2307.09520

Code (1)

shengcheng/aba 공식 구현 pytorch

Tasks

Data Augmentation Domain Generalization Photo to Rest Generalization Single-Source Domain Generalization

Center-aware Adversarial Augmentation for Single Domain Generalization

2023-01-01 · WACV 2023 1 · Tianle Chen, Mahsa Baktashmotlagh, Zijian Wang, Mathieu Salzmann

Domain generalization (DG) aims to learn a model from multiple training (i.e., source) domains that can generalize well to the unseen test (i.e., target) data coming from a different distribution. Single domain generaliz…

Data Augmentation Domain Generalization Photo to Rest Generalization Single-Source Domain Generalization

Adversarial Augmentation Policy Search for Domain and Cross-Lingual Generalization in Reading Comprehension

2020-04-13 · Findings of the Association for Computational Linguistics 2020 · Adyasha Maharana, Mohit Bansal

Reading comprehension models often overfit to nuances of training datasets and fail at adversarial evaluation. Training with adversarially augmented dataset improves robustness against those adversarial attacks but hurts…

Data Augmentation Reading Comprehension

Out-of-Domain Generalization from a Single Source: An Uncertainty Quantification Approach

2021-08-05 · Xi Peng, Fengchun Qiao, Long Zhao

We are concerned with a worst-case scenario in model generalization, in the sense that a model aims to perform well on many unseen domains while there is only one single domain available for training. We propose Meta-Lea…

Domain Generalization Image Classification Meta-Learning Semantic Segmentation+3

CrossMatch: Cross-Classifier Consistency Regularization for Open-Set Single Domain Generalization

2021-09-29 · ICLR 2022 4 · Ronghang Zhu, Sheng Li

Single domain generalization (SDG) is a challenging scenario of domain generalization, where only one source domain is available to train the model. Typical SDG methods are based on the adversarial data augmentation stra…

Data Augmentation Domain Generalization

Generalizing to Unseen Domains via Adversarial Data Augmentation

2018-05-30 · NeurIPS 2018 12 · Riccardo Volpi, Hongseok Namkoong, Ozan Sener, John Duchi 외

We are concerned with learning models that generalize well to different \emph{unseen} domains. We consider a worst-case formulation over data distributions that are near the source domain in the feature space. Only using…

Data Augmentation Semantic Segmentation