Few-Shot Adversarial Domain Adaptation

By Saeid Motiian, Q. Jones, Seyed Mehdi Iranmanesh, Gianfranco Doretto. 2017

435 citations recorded by the source corpus.

This work provides a framework for addressing the problem of supervised domain adaptation with deep models. The main idea is to exploit adversarial learning to learn an embedded subspace that simultaneously maximizes the confusion between two domains while semantically aligning their embedding. The supervised setting becomes attractive especially when there are only a few target data samples that need to be labeled. In this few-shot learning scenario, alignment and separation of semantic probability distributions is difficult because of the lack of data. We found that by carefully designing a training scheme whereby the typical binary adversarial discriminator is augmented to distinguish between four different classes, it is possible to effectively address the supervised adaptation problem. In addition, the approach has a high speed of adaptation, i.e. it requires an extremely low number of labeled target training samples, even one per category can be effective. We then extensively compare this approach to the state of the art in domain adaptation in two experiments: one using datasets for handwritten digit recognition, and one using datasets for visual object recognition.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning

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

In Representation & Self-Supervised Learning

Citations in the corpus

Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning (2021, methodology): “Some domain-adaptive FSL (DA-FSL) methods have been proposed in the case where only a very few samples from the target domain are available (Motiian et al., 2017; Zhao et al., 2020).”
Multiple-Source Adaptation with Domain Classifiers (2020, neutral): “…al., 2008, 2009a, Hoffman et al., 2018, Pan and Yang, 2010, Muandet et al., 2013, Xu et al., 2014, Hoffman et al., 2012, Gong et al., 2013a,b, Zhang et al., 2015, Ganin et al., 2016, Tzeng et al., 2015, Motiian et al., 2017b,a, Wang et al., 2019, Konstantinov and Lampert, 2019, Liu et al., 2015].”
A Theory of Multiple-Source Adaptation with Limited Target Labeled Data (2020, background): “Motiian et al. (2017b), Motiian et al. (2017a) and Wang et al. (2019) proposed to train maximally separated features via adversarial learning.”
Dataset Distillation (2018, methodology): “We also compare our method against a state-of-the-art few-shot domain adaptation method (Motiian et al., 2017).”
Generative Continual Concept Learning (2019, methodology): “This generative strategy has been used in the Machine Learning (ML) literature to address “few-shot learning” (FSL) [11, 12].”
Generalizing from a Few Examples (2019, methodology): “In [95], a variational auto-encoder is first pre-trained from the source tasks, and then cloned to the target task.”
Partial Adversarial Domain Adaptation (2018, background): “Supervised domain adaptation [11,21,22,23] exploits a few labeled data in the target domain.”
Enabling a Single Deep Learning Model for Accurate Gland Instance Segmentation: A Shape-Aware Adversarial Learning Framework (2020, methodology): “For feature alignment, adversarial domain adaptation [28]–[32] has been proved to be an effective technique.”
Target-Independent Domain Adaptation for WBC Classification Using Generative Latent Search (2020, neutral): “2 FADA [56] 60.”
Few-Shot Semantic Segmentation with Prototype Learning (2018, methodology): “In practice, we use the same architecture as regularization [23].”
Unsupervised Multi-Target Domain Adaptation: An Information Theoretic Approach (2018, methodology): “ a labeled source dataset to an unlabeled target domain, where the main strategy is to ﬁnd a feature space such that the confusion between source and target distributions in that space is maximized ( [8, 10, 14, 33, 36, 37, 43, 46]). For this, it is critical to ﬁrst deﬁne a measure of divergence bet”
Rectify Heterogeneous Models with Semantic Mapping (2018, background): “Meta-knowledge also facilitates the cross-task transfer, which is usually used in the few-shot learning (Motiian et al., 2017).”
COVID-19 Image Data Collection: Prospective Predictions Are the Future (2020, background): “, 2020], and domain adaptation [Motiian et al., 2017] will likely be useful in this setting.”
Learning to Diversify for Single Domain Generalization (2021, background): “More recently, fewshot DA [32] is proposed, where only a few labelled target samples together with source samples are available in the training phase.”
Extracting Relationships by Multi-Domain Matching (2018, methodology): “Several other methods use source or target sample generation with GANs on single source domain adaptation [35, 27, 26, 33], but extensions to multi-source domains are not straightforward.”
Deep Cocktail Network: Multi-source Unsupervised Domain Adaptation with Category Shift (2018, background): “Recent relevant researches perform superior in visual recognition cross domain [30] [12] and task [34] and transfer structure learning [4] [21].”
Deep Transfer Learning for Few-Shot SAR Image Classification (2019, background): “This setting is called semi-supervised 99 domain adaptation in the machine learning literature [14].”