Deep Domain-Adversarial Image Generation for Domain Generalisation
By Kaiyang Zhou, Yongxin Yang, Timothy M. Hospedales, Tao A. Xiang. AAAI Conference on Artificial Intelligence, 2020
500 citations recorded by the source corpus.
Machine learning models typically suffer from the domain shift problem when trained on a source dataset and evaluated on a target dataset of different distribution. To overcome this problem, domain generalisation (DG) methods aim to leverage data from multiple source domains so that a trained model can generalise to unseen domains. In this paper, we propose a novel DG approach based on Deep Domain-Adversarial Image Generation (DDAIG). Specifically, DDAIG consists of three components, namely a label classifier, a domain classifier and a domain transformation network (DoTNet). The goal for DoTNet is to map the source training data to unseen domains. This is achieved by having a learning objective formulated to ensure that the generated data can be correctly classified by the label classifier while fooling the domain classifier. By augmenting the source training data with the generated unseen domain data, we can make the label classifier more robust to unknown domain changes. Extensive experiments on four DG datasets demonstrate the effectiveness of our approach.
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/14d9d5fd85641f91b1c6364fd1206e88d9252ae0
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
Citations in the corpus
- Domain Generalization: A Survey (2021, methodology): “Following [43], Khosla et al. [46] targeted the cross-dataset generalization problem in classification and detection tasks, and proposed to learn domain-specific bias vectors and domain-agnostic weight vectors based on support vector machine (SVM) classifiers.”
- Domain Generalization with MixStyle (2021, methodology): “Whereas L2A-OT works at the pixel level: it trains an image generator by maximizing the domain difference (measured by optimal transport) between the original and the generated images, which introduces much heavier computational overhead than MixStyle in terms of GPU memory and training time.”
- Deep Stable Learning for Out-Of-Distribution Generalization (2021, background): “Several works propose to enlarge the available data space with augmentation of source domains [6, 49, 59, 44, 64, 63].”
- Single-Source Domain Expansion Network for Cross-Scene Hyperspectral Image Classification (2022, methodology): “adversarially trained a transformation network for data augmentation instead of directly updating the inputs by gradient ascent [17].”
- Generalizing to Unseen Domains: A Survey on Domain Generalization (2021, background): “[37] adversarially trained a transformation network for data augmentation instead of directly updating the inputs by”
- Domain Generalization using Causal Matching (2020, methodology): “On the PACS dataset with ResNet-18 architecture (Table 4), our methods are competitive to state-of-the-Domain (Carlucci et al., 2019), DDAIG (Zhou et al., 2020), SagNet (Nam et al., 2019), DDEC (Asadi et al., 2019), were taken from the Do-mainBed (Gulrajani & Lopez-Paz, 2020) paper.”
- In Search of Lost Domain Generalization (2020, background): “This augmentation protocol is increasingly standard in state-of-the-art DG work (Nam et al., 2019; Huang et al., 2020; Krueger et al., 2020; Carlucci et al., 2019a; Zhou et al., 2020; Dou et al., 2019; Hendrycks et al., 2020; Wang et al., 2020a; Seo et al., 2020; Chattopadhyay et al., 2020).”
- A Fourier-based Framework for Domain Generalization (2021, methodology): “Compared with the SOTA, our FACT clearly beats the methods based on adversarial data augmentation or meta learning, including the latest MASF [4], DDAIG [50] and L2A-OT [51], yet FACT enjoys an efficient training process without any additional adversarial or episodic training steps.”