Partial Adversarial Domain Adaptation

By Zhangjie Cao, Li-jie Ma, Mingsheng Long, Jianmin Wang. 2018

491 citations recorded by the source corpus.

Domain adversarial learning aligns the feature distributions across the source and target domains in a two-player minimax game. Existing domain adversarial networks generally assume identical label space across different domains. In the presence of big data, there is strong motivation of transferring deep models from existing big domains to unknown small domains. This paper introduces partial domain adaptation as a new domain adaptation scenario, which relaxes the fully shared label space assumption to that the source label space subsumes the target label space. Previous methods typically match the whole source domain to the target domain, which are vulnerable to negative transfer for the partial domain adaptation problem due to the large mismatch between label spaces. We present Partial Adversarial Domain Adaptation (PADA), which simultaneously alleviates negative transfer by down-weighing the data of outlier source classes for training both source classifier and domain adversary, and promotes positive transfer by matching the feature distributions in the shared label space. Experiments show that PADA exceeds state-of-the-art results for partial domain adaptation tasks on several datasets.

Subfields: Artificial Intelligence, Infectious Diseases

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 Infectious Diseases

Citations in the corpus

Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation (2022, methodology): “To alleviate the negative transfer caused by source private classes, existing PDA methods [1,2,16] decrease the transferability weights of source pri-”
Separate to Adapt: Open Set Domain Adaptation via Progressive Separation (2019, methodology): “Existing domain adaptation methods seek to bridge domain gap by distribution matching in feature-level [36, 4, 17, 19, 35, 3] or in pixellevel [10, 31, 11, 22, 16].”
Universal Source-Free Domain Adaptation (2020, methodology): “Further, we aim to exploit the learned generative classifier from the Procurement stage to complement for the purpose of separate ad-hoc networks (critic or discriminator) as utilized by the prior works [52, 6].”
A Survey on Negative Transfer (2020, background): “If the source and target domains have different label spaces, then a TL algorithm that transfers the outlier source classes to the target domain would mis-align the features of outlier source classes to certain target classes, migrating harmful knowledge from the source domain to the target domain a”
Attending to Discriminative Certainty for Domain Adaptation (2019, background): “Some works closely related to MADA have been proposed in [4, 3].”
Open Compound Domain Adaptation (2019, background): “Another route of research aims to tackle the category sharing/unsharing issues between source and target domain, namely open set [37, 43] and partial set [55, 3] domain adaptation.”
Domain Consensus Clustering for Universal Domain Adaptation (2021, methodology): “For example, on Ofﬁce-Home, our method (‘U’) achieves 70.9% average accuracy, which outperforms PADA [4] (62.1”
Multiple Class Novelty Detection Under Data Distribution Shift (2020, background): “These extensions include partial domain adaptation [5], universal domain adaptation [53] and open-set domain adaptation [36].”
Characterizing and Avoiding Negative Transfer (2018, methodology): “a RevGrad [11], ADDA [29], PADA [4], GTA [26].”
Domain Generalization: A Survey (2021, unknown): “DG, such as single- [150] and multi-source [60] DA, and heterogeneous DA [17], [156], [157], [158].”
Learning to Transfer Examples for Partial Domain Adaptation (2019, methodology): “PDA enables knowledge transfer from a big domain of many labels to a small domain of few labels.”
Minimum Class Confusion for Versatile Domain Adaptation (2019, methodology): “For a fair comparison, we follow the protocol of PADA [3] and AFN [49], where the first 25 categories (in alphabetic order) of the Office-Home dataset are taken as the target domain.”
Domain2Vec: Domain Embedding for Unsupervised Domain Adaptation (2020, methodology): “8 PADA [67] 34.”
Applications of Unsupervised Deep Transfer Learning to Intelligent Fault Diagnosis: A Survey and Comparative Study (2019, methodology): “2) Partial adversarial domain adaptation (PADA): One of popular partial UDTL methods named partial adversarial domain adaptation (PADA) was proposed by Cao et. al [136].”
Domain Generalization by Solving Jigsaw Puzzles (2019, background): “To get closer to real world conditions, recent work has started to focus on cases where the source data are drawn from multiple distributions [30, 48] and the target covers only a part of the source classes [5, 1].”