Corpus ID: 236469141

Adversarial Unsupervised Domain Adaptation with Conditional and Label Shift: Infer, Align and Iterate

@article{Liu2021AdversarialUD,
  title={Adversarial Unsupervised Domain Adaptation with Conditional and Label Shift: Infer, Align and Iterate},
  author={Xiaofeng Liu and Zhenhua Guo and Site Li and Fangxu Xing and Jane Jia You and C.-C. Jay Kuo and Georges El Fakhri and Jonghye Woo},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.13469}
}
  • Xiaofeng Liu, Zhenhua Guo, +5 authors Jonghye Woo
  • Published 2021
  • Computer Science
  • ArXiv
In this work, we propose an adversarial unsupervised domain adaptation (UDA) approach with the inherent conditional and label shifts, in which we aim to align the distributions w.r.t. both p(x|y) and p(y). Since the label is inaccessible in the target domain, the conventional adversarial UDA assumes p(y) is invariant across domains, and relies on aligning p(x) as an alternative to the p(x|y) alignment. To address this, we provide a thorough theoretical and empirical analysis of the conventional… Expand

Figures and Tables from this paper

Partial Domain Adaptation without Domain Alignment
  • Weikai Li, Songcan Chen
  • Computer Science
  • ArXiv
  • 2021
TLDR
This is the first naive attempt to address the PDA without domain alignment, and instantiate the model smoothness as a quite simple intra-domain structure preserving (IDSP) to enhance the adaptability of the model. Expand
Recursively Conditional Gaussian for Ordinal Unsupervised Domain Adaptation
TLDR
A recursively conditional Gaussian (RCG) set is proposed for ordered constraint modeling, which admits a tractable joint distribution prior to the latent space and is able to control the density of content vectors that violate the poset constraint by a simple “three-sigma rule”. Expand

References

SHOWING 1-10 OF 70 REFERENCES
Domain Generalization under Conditional and Label Shifts via Variational Bayesian Inference
  • Xiaofeng Liu, Bo Hu, +6 authors Jonghye Woo
  • Computer Science
  • IJCAI
  • 2021
TLDR
This work proposes a novel variational Bayesian inference framework to enforce the conditional distribution alignment w.r.t. p(x|y) via the prior distribution matching in a latent space, which also takes the marginal label shift p(y) into consideration with the posterior alignment. Expand
Re-weighted Adversarial Adaptation Network for Unsupervised Domain Adaptation
TLDR
The Re-weighted Adversarial Adaptation Network (RAAN) is proposed to reduce the feature distribution divergence and adapt the classifier when domain discrepancies are disparate and to match the label distribution and embed it into the adversarial training. Expand
Generate to Adapt: Aligning Domains Using Generative Adversarial Networks
TLDR
This work proposes an approach that leverages unsupervised data to bring the source and target distributions closer in a learned joint feature space by inducing a symbiotic relationship between the learned embedding and a generative adversarial network. Expand
PANDA: Prototypical Unsupervised Domain Adaptation
TLDR
This work attempts to calibrate the noisy pseudo labels with prototypes, and develops an unified framework termed PrototypicAl uNsupervised Domain Adaptation (PANDA) for UDA, which achieves state-of-the-art or competitive results on multiple UDA benchmarks including both object recognition and semantic segmentation tasks. Expand
Adversarial Discriminative Domain Adaptation
TLDR
It is shown that ADDA is more effective yet considerably simpler than competing domain-adversarial methods, and the promise of the approach is demonstrated by exceeding state-of-the-art unsupervised adaptation results on standard domain adaptation tasks as well as a difficult cross-modality object classification task. Expand
An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation
TLDR
This work proposes to explicitly train a domain-invariant classifier by generating and defensing against pointwise feature space adversarial perturbations and achieves the state-of-the-art performance on two challenging domain adaptation tasks for semantic segmentation. Expand
Adversarial Dropout Regularization
TLDR
This work proposes a novel approach, Adversarial Dropout Regularization (ADR), to encourage the generator to output more discriminative features for the target domain, using dropout on the classifier network. Expand
Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment
TLDR
Asymmetrically-relaxed distribution alignment is proposed, a new approach that overcomes some limitations of standard domain-adversarial algorithms and characterize precise assumptions under which the algorithm is theoretically principled and demonstrate empirical benefits on both synthetic and real datasets. Expand
Gotta Adapt 'Em All: Joint Pixel and Feature-Level Domain Adaptation for Recognition in the Wild
TLDR
This work uses 3D geometry and image synthesis based on a generalized appearance flow to preserve identity across pose transformations, while using an attribute-conditioned CycleGAN to translate a single source into multiple target images that differ in lower-level properties such as lighting. Expand
CyCADA: Cycle-Consistent Adversarial Domain Adaptation
TLDR
A novel discriminatively-trained Cycle-Consistent Adversarial Domain Adaptation model that adapts representations at both the pixel-level and feature-level, enforces cycle-consistency while leveraging a task loss, and does not require aligned pairs is proposed. Expand
...
1
2
3
4
5
...