domain adaptation论文记录

1 An Intelligent Fault Diagnosis Method Based on Domain Adaptation and Its Application for Bearings Under Polytropic Working Conditions

创新点:
(1) a domain adaptation method based on multi-scale mixed domain feature is proposed.
(2) adaptive domain adaptation(同时考虑条件概率和边缘概率)
(3)A feature optimizer based on GS_XGBoost is proposed, which not only reduces the dimension of features and eliminates redundant information, but also gives the importance ranking of features to understand the contribution of different features for diagnosis and recognition, and then selects sensitive feature subsets to diagnosis。

2 Deep Learning-Based Partial Domain Adaptation Method on Intelligent Machinery Fault Diagnostics

创新点
(1) address the partial domain adaptation
(2) The conditional data alignment and unsupervised prediction consistency schemes are proposed to achieve partial domain adaptation.

3 Adversarial domain-invariant generalization a generic domain-regressive framework for bearing fault diagnosis under unseen conditions

创新点：
(1) 提出了domain generalization解决unseen target问题

(2)Customized IN and SN strategies achieve cross-domain feature normalization to promote domain generalizability
(3)an adaptive weight strategy achieves weight self-learning during multitask learning to improve performance.(自适应调整各损失的相对权重，解决手动调参的问题)

4 Adversarial Multiple-Target Domain Adaptation for Fault Classification（开源）

创新点：
(1) 解决单源域多目标域问题

5 Adversarial Entropy Optimization for Unsupervised Domain Adaptation

创新点

当特征来自于源域或目标域的独立分布时，最小化熵损失，当特征来自于目标域和源域的组合分布时，最大化熵损失。

6 Deep Coupled Joint Distribution Adaptation Network: A Method for Intelligent Fault Diagnosis Between Artificial and Real Damages

创新点：

(1)在提取特征时，源域和目标域的网络结构不一样，用一个正则项来联系两个网络；
(2)同时考虑条件概率分布和边缘概率分布。

7 Conditional Adversarial Domain Adaptation With Discrimination Embedding for Locomotive Fault Diagnosis

创新点：

(1)Instead of inputting features of the source and target domains into the domain classifier, the multilinear map of features and label predictions are inputted into the domain classifier
(2)CADA

8 Extreme Learning Machine Based on Maximum Weighted Mean Discrepancy for Unsupervised Domain Adaptation

创新点：

(1)考虑每一个样本的权重；
(2)引入超极限学习

9 Adaptive Graph Adversarial Networks for Partial Domain Adaptation

创新点：
(1)针对partial domain adaptation问题
(2)we propose Adaptive Graph Adversarial Networks (AGAN) consisting of two specialized modules. The adaptive class-relational graph module is designed to utilize the intra-and inter-domain structures through adaptive feature propagation. Complementarily, the sample-level commonness predictor computes a commonness score of each sample.