Abstract:An important challenge in unsupervised domain adaptation (UDA) is how to capture domain-specific discriminative information in the target domain data. For the labels in the target domain are unavailable, some methods uncover the intrinsic discriminative information by clustering the target domain features in the embedding space. However, most of existing works only consider filtering reliable target domain samples by setting a fixed distance threshold, ignoring the potential intraclass and interclass relationships and cannot set appropriate distance thresholds for different transfer tasks and different classes. To this end, we propose a new pseudolabeling framework named the dynamic distance threshold learning framework (DDT) for UDA. In our framework, we construct distance memory module to encode intraclass and interclass distribution information, and select reliable target domain samples satisfying intraclass compactness and interclass independence via dynamic intraclass and interclass distance thresholds. Specifically, we define class centroid memory modules, intraclass and interclass distance memory modules in the embedding space. Then, we retrieve the intraclass compactness distance threshold and interclass independence distance threshold from the distance memory based on the class and classpair information of target domain sample. Finally, we use reliable target domain samples to update the class centroid and the distance memory module, so that the distance memory module is dynamically updated following the model training. Benefiting from DDT, the UDA model can be updated with reliable target domain samples for training, and the joint distribution can be further aligned at the category level. As a lightweight technique, DDT constructs class center and distance memory modules to transfer the intraclass and interclass distribution relationships from the source domain to the target domain, and sets dynamic intraclass and interclass distances for different classes in different transfer tasks. Without any additional training parameters, the framework can be easily plugged into various domain adaptation paradigms. Experiments on benchmark datasets show that our method can filter the noise in pseudolabels and confers a competitive advantage compared with existing state-of-the-art domain adaptation methods.

CMFT: Contrastive Memory Feature Transfer for Non-shared-and-Imbalanced Unsupervised Domain Adaption

CMFT: Contrastive Memory Feature Transfer for Nonshared-and-Imbalanced Unsupervised Domain Adaption

Unsupervised Domain Adaptation With Class-Aware Memory Alignment

NI-UDA: Graph Contrastive Domain Adaptation for Nonshared-and-Imbalanced Unsupervised Domain Adaptation

Unsupervised Multi-Target Domain Adaptation Incremental Method Based on Contrastive Learning

Unsupervised domain adaptation via coarse-to-fine feature alignment method using contrastive learning

Balancing Transferability and Discriminability for Unsupervised Domain Adaptation.

Domain consensual contrastive learning for few-shot universal domain adaptation

Transferable Discriminative Feature Mining for Unsupervised Domain Adaptation.

ConFUDA: Contrastive Fewshot Unsupervised Domain Adaptation for Medical Image Segmentation

Discriminative Feature Mining and Alignment for Unsupervised Domain Adaptation

Unsupervised Domain Adaptation by Cross-Prototype Contrastive Learning for Medical Image Segmentation.

Unsupervised Domain Adaptation Via Reliable Pseudolabeling Based Memory Module and Dynamic Distance Threshold Learning

NI-UDA: Graph Adversarial Domain Adaptation from Non-shared-and-Imbalanced Big Data to Small Imbalanced Applications

A Focally Discriminative Loss for Unsupervised Domain Adaptation

Integrating multimodal contrastive learning with prototypical domain alignment for unsupervised domain adaptation of time series

Bilateral Co-Transfer for Unsupervised Domain Adaptation

Memory-Assisted Sub-Prototype Mining for Universal Domain Adaptation

Joint Clustering and Discriminative Feature Alignment for Unsupervised Domain Adaptation

Unsupervised domain adaptation via discriminative feature learning and classifier adaptation from center-based distances

Contrastive Domain Adaptation for Time-Series via Temporal Mixup