Unsupervised Domain Adaptation with Similarity Learning

By Pedro H. O. Pinheiro. 2017

294 citations recorded by the source corpus.

The objective of unsupervised domain adaptation is to leverage features from a labeled source domain and learn a classifier for an unlabeled target domain, with a similar but different data distribution. Most deep learning approaches to domain adaptation consist of two steps: (i) learn features that preserve a low risk on labeled samples (source domain) and (ii) make the features from both domains to be as indistinguishable as possible, so that a classifier trained on the source can also be applied on the target domain. In general, the classifiers in step (i) consist of fully-connected layers applied directly on the indistinguishable features learned in (ii). In this paper, we propose a different way to do the classification, using similarity learning. The proposed method learns a pairwise similarity function in which classification can be performed by computing similarity between prototype representations of each category. The domain-invariant features and the categorical prototype representations are learned jointly and in an end-to-end fashion. At inference time, images from the target domain are compared to the prototypes and the label associated with the one that best matches the image is outputed. The approach is simple, scalable and effective. We show that our model achieves state-of-the-art performance in different unsupervised domain adaptation scenarios.

Subfields: Artificial Intelligence, Cancer Research, Computer Vision and Pattern Recognition

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 Cancer Research

In Computer Vision and Pattern Recognition

Citations in the corpus

Continual Semantic Segmentation via Repulsion-Attraction of Sparse and Disentangled Latent Representations (2021, methodology): “Prototypes-based regularizing terms gained a great interest and, in particular, have been largely used in the literature of few-shot learning [15, 63, 60], to learn prototypical representations of each category, and domain adaptation, to enforce orthogonality [50, 65] or centroid matching [70, 12].”
Category Anchor-Guided Unsupervised Domain Adaptation for Semantic Segmentation (2019, methodology): “To overcome this issue, a variety of domain adaptation methods for classiﬁcation [6, 23, 37, 28, 41, 3, 19], detection [38, 16], and segmentation [7, 14, 13, 27, 34, 40, 21, 47] have been proposed.”
Sequential Unsupervised Domain Adaptation through Prototypical Distributions (2020, background): “Figure 2 represents the testing splits of the source and the target domains and samples of the prototypical for the S → M Task JAN [12] DJT [21] GtA [6] SimNet [39] CDAN [9] MCD [40] Ours Syn.”
HGNet: Hybrid Generative Network for Zero-Shot Domain Adaptation (2020, background): “To learn transferable information, it assumes that instances of target modality are available [23, 17, 5].”
Unsupervised Domain Adaptation via Structurally Regularized Deep Clustering (2020, background): “A typical line of works [16, 43, 53, 63] leverages a domain-adversarial task to align the source and target domains as a whole so that class labels can be transferred from the source domain to the unlabeled target one.”
Model Adaptation: Unsupervised Domain Adaptation Without Source Data (2020, background): “The accuracy is increased to 83.3%, which surpasses SimDA [45] with ResNet152 by 10.4%.”
Stochastic Classifiers for Unsupervised Domain Adaptation (2020, unknown): “4 SimNet [38] - 96.”
Unsupervised Domain Adaptation in Semantic Segmentation via Orthogonal and Clustered Embeddings (2020, background): “The orthogonality between feature vectors has been recently proposed for UDA in [38, 65].”
Multi-Adversarial Discriminative Deep Domain Generalization for Face Presentation Attack Detection (2019, methodology): “The straightforward way to solve this problem is to exploit the domain adaptation technique [27, 12, 25, 20, 34, 7, 24, 29, 6, 30, 33, 3] to align the feature distribution between training and testing data so that the trained model with source data can be adapted on the target data.”
Robust Optimal Transport with Applications in Generative Modeling and Domain Adaptation (2020, background): “Outliers are assigned smaller weights compared to the in-distribution samples, and Source Only 50.7 DAN [14] 53.0 RTN [16] 53.6 DANN [10] 55.0 JAN-A [17] 61.6 GTA [25] 69.5 SimNet [22] 69.6 CDAN-E [15] 70.0 there is a clear separation between their corresponding histograms.”
Gradually Vanishing Bridge for Adversarial Domain Adaptation (2020, methodology): “We compare our methods with state-of-art domain adaptation methods: DAN [23], DANN [9], ADDA [40], GTA [36], CDAN [24], Simnet [32], MCD [35] and MDD [47].”
Towards Discriminability and Diversity: Batch Nuclear-Norm Maximization Under Label Insufficient Situations (2020, unknown): “6 Simnet [36] 85.”
A Survey of Unsupervised Deep Domain Adaptation (2018, methodology): “not BN French et al.[68] En,N sq. diff. ✓ EMA Co-DA[124]a DI,En,N,TD L1 ✓ ✓ ✓ optional VADA[220]a DI,TD ✓ ✓ ✓ DeepJDOT[50] DI JDOT ✓ ✓ CyCADA[96] DI,DM ✓ ✓ ✓ ✓ ✓ ✓ Gen. to Adapt[208] DI ✓ ✓ ✓ ✓ SimNet[187] DI prototypes ✓ MADA[183] DI,En ✓ ✓ ✓ MCD[206] DI,En,TD ✓ ✓ ✓ ✓ GAGL[250] DI,TD ✓ ✓ ✓ ✓ ✓ SBAD”
Learning Semantic Representations for Unsupervised Domain Adaptation (2018, background): “(Pinheiro, 2017) classifies the input samples by computing the distances between prototype representations of each category.”
Improving Domain-Specific Classification by Collaborative Learning with Adaptation Networks (2019, unknown): “In (Pinheiro 2018), Pinheiro used an adversarial loss to learn domain-invariant features, and simultaneously train a similarity-based classiﬁer by measuring the similarity of each sample to categorical prototypes.”