Domain-Adaptive Few-Shot Learning

@article{Zhao2021DomainAdaptiveFL,
  title={Domain-Adaptive Few-Shot Learning},
  author={An Zhao and Mingyu Ding and Zhiwu Lu and Tao Xiang and Yulei Niu and Jiechao Guan and Ji-rong Wen and Ping Luo},
  journal={2021 IEEE Winter Conference on Applications of Computer Vision (WACV)},
  year={2021},
  pages={1389-1398}
}
  • An Zhao, Mingyu Ding, Ping Luo
  • Published 19 March 2020
  • Computer Science
  • 2021 IEEE Winter Conference on Applications of Computer Vision (WACV)
Existing few-shot learning (FSL) methods make the implicit assumption that the few target class samples are from the same domain as the source class samples. However, in practice, this assumption is often invalid –the target classes could come from a different domain. This poses an additional challenge of domain adaptation (DA) with few training samples. In this paper, the problem of domain-adaptive few-shot learning (DA-FSL) is tackled, which is expected to have wide use in real-world… 

Figures and Tables from this paper

Revisiting Mid-Level Patterns for Cross-Domain Few-Shot Recognition
TLDR
A challenging subset of CDFSL where the novel classes are in distant domains from base classes is studied, by revisiting the mid-level features, which are more transferable yet under-explored in main stream FSL work, and a residual-prediction task is proposed to boost the discriminability of mid- level features.
Revisiting Mid-Level Patterns for Distant-Domain Few-Shot Recognition
TLDR
This paper revisits distant-domain FSL, a challenging subset of CDFSL, by revisiting the mid-level features, which are more transferable yet under-explored in main stream FSL work, and proposes a residual-prediction task to boost the discriminability of mid- level features.
TOHAN: A One-step Approach towards Few-shot Hypothesis Adaptation
TLDR
A targetoriented hypothesis adaptation network (TOHAN) is proposed to solve the FHA problem, where highly-compatible unlabeled data is generated to help train a target-domain classifier and the private information in SD will be protected well.
Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning
TLDR
This paper considers the novel problem of repurposing pretrained MAML checkpoints to solve new few-shot classification tasks and proposes an alternative meta-testing procedure that outperforms “vanilla” MAMl on same-domain and cross-domains benchmarks using both SGD and Adam optimizers and shows improved robustness to the choice of base stepsize.
PICA: Point-wise Instance and Centroid Alignment Based Few-shot Domain Adaptive Object Detection with Loose Annotations
TLDR
This work proposes a method that extends the point-wise alignment from classification to object detection and is not only robust against label noises of background ROIs but also robust against the scarcity of labeled target objects.
Bridging Few-Shot Learning and Adaptation: New Challenges of Support-Query Shift
TLDR
This work addresses the new and challenging problem of Few-Shot Learning under Support/Query Shift (FSQS) i.e., when support and query instances are sampled from related but different distributions, and studies both the role of Batch-Normalization and Optimal Transport in aligning distributions.
PGADA: Perturbation-Guided Adversarial Alignment for Few-shot Learning Under the Support-Query Shift
TLDR
A novel adversarial data augmentation method, namely Perturbation-Guided Adversarial Alignment (PGADA) is proposed, which generates the hard examples in a self-supervised manner and introduces Regularized Optimal Transportation to derive a smooth optimal transportation plan.
Few-Shot Domain Adaptation For End-to-End Communication
TLDR
This work proposes to address the problem of end-to-end learning of a communication system using an autoencoder using a fast and sample-efficient (few-shot) domain adaptation method that does not change the encoder and decoder networks.
Unsupervised Domain Adaptive 3D Detection with Multi-Level Consistency
TLDR
A novel and unified framework, Multi-Level Consistency Network (MLC-Net), which employs a teacher-student paradigm to generate adaptive and reliable pseudo-targets and out-performs existing state-of-the-art methods on standard benchmarks.
Few-Shot Domain Adaptation for Low Light RAW Image Enhancement
TLDR
This work presents a novel few-shot domain adaptation method to utilize the existing source camera labeled data with few labeled samples from the target camera to improve the target domain’s enhancement quality in extreme low-light imaging.
...
...

References

SHOWING 1-10 OF 74 REFERENCES
Few-Shot Adversarial Domain Adaptation
TLDR
This work provides a framework for addressing the problem of supervised domain adaptation with deep models by carefully designing a training scheme whereby the typical binary adversarial discriminator is augmented to distinguish between four different classes.
Unsupervised Domain Adaptation for Zero-Shot Learning
TLDR
A novel ZSL method is proposed based on unsupervised domain adaptation which uses the target domain class labels' projections in the semantic space to regularise the learned target domain projection thus effectively overcoming the projection domain shift problem.
Unsupervised Domain Adaptation: An Adaptive Feature Norm Approach
TLDR
This paper empirically reveal that the erratic discrimination of target domain mainly reflects in its much lower feature norm value with respect to that of the source domain, and demonstrates that adapting feature norms of source and target domains to achieve equilibrium over a large range of values can result in significant domain transfer gains.
Semi-Supervised Domain Adaptation via Minimax Entropy
TLDR
A novel Minimax Entropy (MME) approach that adversarially optimizes an adaptive few-shot model for semi-supervised domain adaptation (SSDA) setting, setting a new state of the art for SSDA.
Large-Scale Few-Shot Learning: Knowledge Transfer With Class Hierarchy
TLDR
This work proposes a novel large-scale FSL model by learning transferable visual features with the class hierarchy which encodes the semantic relations between source and target classes and significantly outperforms not only the NN baseline but also the state-of-the-art alternatives.
M-ADDA: Unsupervised Domain Adaptation with Deep Metric Learning
TLDR
It is shown that M-ADDA performs significantly better on the digits adaptation datasets of MNIST and USPS, suggesting that using metric-learning for domain adaptation can lead to large improvements in classification accuracy for the domain adaptation task.
Meta-Transfer Learning for Few-Shot Learning
TLDR
A novel few-shot learning method called meta-transfer learning (MTL) which learns to adapt a deep NN for few shot learning tasks and introduces the hard task (HT) meta-batch scheme as an effective learning curriculum for MTL.
Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformation
TLDR
The core idea is to use feature-wise transformation layers for augmenting the image features using affine transforms to simulate various feature distributions under different domains in the training stage, and applies a learning-to-learn approach to search for the hyper-parameters of the feature- wise transformation layers.
Beyond Sharing Weights for Deep Domain Adaptation
TLDR
This work introduces a two-stream architecture, where one operates in the source domain and the other in the target domain, and demonstrates that this both yields higher accuracy than state-of-the-art methods on several object recognition and detection tasks and consistently outperforms networks with shared weights in both supervised and unsupervised settings.
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.
...
...