May 22, 2023 · We study a more practical SF-UDA task, termed imbalance-agnostic SF-UDA, where the class distributions of both the unseen source domain and unlabeled target ...
Sep 9, 2024 · Source-free Unsupervised Domain Adaptation (SF-UDA) aims to adapt a well-trained source model to an unlabeled target domain without access ...
To address this issue, we study a more practical SF-UDA task, termed imbalance-agnostic SF-UDA, where the class distributions of both the unseen source domain ...
A new robust contrastive prototype adaptation strategy to align each pseudo-labeled target data to the corresponding source prototypes via contrastive learning ...
Co-authors ; Imbalance-Agnostic Source-Free Domain Adaptation via Avatar Prototype Alignment. H Lin, M Tan, Y Zhang, Z Qiu, S Niu, D Liu, Q Du, Y Liu. arXiv ...
Imbalanced Source-free Domain Adaptation - Semantic Scholar
www.semanticscholar.org › paper › Imba...
Imbalanced Source-free Domain Adaptation is presented, which first train a uniformed model from the source domain, and then proposes secondary label ...
To address this issue, we study a more practical SF-UDA task, termed imbalance-agnostic SF-UDA, where the class distributions of both the unseen source domain ...
People also ask
What is source free domain adaptation?
What is the domain adaptation feature?
What is domain adaptation in NLP?
What is universal domain adaptation?
Source-free Domain Adaptation via Avatar Prototype Generation ... - IJCAI
www.ijcai.org › proceedings
We study a practical domain adaptation task, called source-free unsupervised domain adaptation (UDA) problem, in which we cannot access source domain data ...
Missing: Imbalance- Agnostic
Specifically, for better prototype adaptation in the imbalance-agnostic scenario, T-CPGA applies a new pseudo label generation strategy to identify unknown ...
Sep 7, 2024 · Our approach is robust to differences in the source and target label distributions and thus applicable to both balanced and imbalanced domain ...