Domain Adaptation Network Research Articles

Prototypical Contrastive Domain Adaptation Network for Nonstationary EEG Classification

Prototypical Contrastive Domain Adaptation Network for Nonstationary EEG Classification

Towards JPEG-Resistant Image Forgery Detection and Localization Via Self-Supervised Domain Adaptation.

With wide applications of image editing tools, forged images (splicing, copy-move, removal and etc.) have been becoming great public concerns. Although existing image forgery localization methods could achieve fairly good results on several public datasets, most of them perform poorly when the forged images are JPEG compressed as they are usually done in social networks. To tackle this issue, in this paper, a self-supervised domain adaptation network, which is composed of a backbone network with Siamese architecture and a compression approximation network (ComNet), is proposed for JPEG-resistant image forgery detection and localization. To improve the performance against JPEG compression, ComNet is customized to approximate the JPEG compression operation through self-supervised learning, generating JPEG-agent images with general JPEG compression characteristics. The backbone network is then trained with domain adaptation strategy to localize the tampering boundary and region, and alleviate the domain shift between uncompressed and JPEG-agent images. Extensive experimental results on several public datasets show that the proposed method outperforms or rivals to other state-of-the-art methods in image forgery detection and localization, especially for JPEG compression with unknown QFs.

Intelligent Fault Diagnosis of Bearings in Unsupervised Dynamic Domain Adaptation Networks Under Variable Conditions

Intelligent Fault Diagnosis of Bearings in Unsupervised Dynamic Domain Adaptation Networks Under Variable Conditions

CLIPREC: Graph-Based Domain Adaptive Network for Zero-Shot Referring Expression Comprehension

CLIPREC: Graph-Based Domain Adaptive Network for Zero-Shot Referring Expression Comprehension

Adversarial domain adaptation network with MixMatch for incipient fault diagnosis of PMSM under multiple working conditions

The incipient faults diagnosis under multiple working conditions has been a challenging problem due to the small difference of incipient fault features, and different feature divergences and distributions under different working conditions. In this paper, an Adversarial Domain Adaptation Network with MixMatch (ADANM) is proposed for the incipient faults diagnosis of permanent magnet synchronous motor (PMSM) under multiple working conditions. In ADANM, through selecting multi-scale convolutional kernels, an Adaptive Multi-scale Convolutional Neural Network (AM-CNN) is designed to improve the adaptability of extracted features for the incipient faults of PMSM under different working conditions. Meanwhile, the MixMatch and confidence-thresholding embedded in the adversarial network are employed to reduce the conditional distribution discrepancy by reducing the number of false pseudo-labels. The proposed ADANM is evaluated by diagnosing 4 types of incipient faults in PMSM under multiple working conditions. Extensive experiments are conducted on the hardware-in-the-loop (HIL) real-time platform, and our method outperforms existing methods by an average accuracy increase of approximately 8.12% across three different datasets. This significant improvement validates the effectiveness and superiority of the proposed method.

Deep adversarial reconstruction classification network for unsupervised domain adaptation

Deep adversarial reconstruction classification network for unsupervised domain adaptation

Multi-condition tool wear prediction for milling CFRP base on a novel hybrid monitoring method

In the carbon fiber-reinforced plastic milling process, the high abrasive property of carbon fiber will lead to the rapid growth of tool wear, resulting in poor surface quality of parts. However, due to the signal data distribution discrepancy under different working conditions, addressing the problem of local degradation and low prediction accuracy in tool wear monitoring model is a significant challenge. This paper proposes an entropy criterion deep conditional domain adaptation network, which effectively exploits domain invariant features of the signals and enhances the stability of model training. Furthermore, a novel unsupervised optimization method based on tool wear distribution is proposed, which refines the monitoring results of data-driven models. This approach reduces misclassification of tool wear conditions resulting from defects in data-driven models and interference from the manufacturing process, thereby enhancing the accuracy of the monitoring model. The experimental results show that the hybrid method provides assurance for the accurate construction of tool wear monitoring model under different working conditions.

A fine-grained feature decoupling based multi-source domain adaptation network for rotating machinery fault diagnosis

Multi-source domain adaptation, an effective solution for rotating machinery fault diagnosis, has achieved great success. However, previous multi-source domain adaptation based methods typically achieve domain consistency by aligning the global distribution of the source and target domains, regardless of private domain information and sufficient feature transferability, thereby limiting the diagnosis performance in cross-domain diagnostic scenarios. To address these issues, we propose a novel fine-grained feature decoupling based multi-source domain adaptation network (FD-MDAN) for fault diagnosis. Leveraging fine-grained feature decoupling, our FD-MDAN weakens specific features with private domain information and reinforces invariant features with diverse transferability at both of domain and category levels. We further develop a diversity regular term to minimize the feature redundancy, enhancing the transferability. In addition, to ensure that these decoupled features retain representative information of the original data, a reconstruction module is designed to make the reconstructed data and the original data consistent. Extensive experiments are conducted on three rotating machinery datasets. Benefiting from these components, our proposed FD-MDAN demonstrates state-of-the-art performance on various transfer diagnosis tasks.

Multiweight Adversarial Open-Set Domain Adaptation Network for Machinery Fault Diagnosis With Unknown Faults

Multiweight Adversarial Open-Set Domain Adaptation Network for Machinery Fault Diagnosis With Unknown Faults

Training multi-source domain adaptation network by mutual information estimation and minimization

We address the problem of Multi-Source Domain Adaptation (MSDA), which trains a neural network using multiple labeled source datasets and an unlabeled target dataset, and expects the trained network to well classify the unlabeled target data. The main challenge in this problem is that the datasets are generated by relevant but different joint distributions. In this paper, we propose to address this challenge by estimating and minimizing the mutual information in the network latent feature space, which leads to the alignment of the source joint distributions and target joint distribution simultaneously. Here, the estimation of the mutual information is formulated into a convex optimization problem, such that the global optimal solution can be easily found. We conduct experiments on several public datasets, and show that our algorithm statistically outperforms its competitors. Video and code are available at https://github.com/sentaochen/Mutual-Information-Estimation-and-Minimization.

Learning robust features alignment for cross-domain medical image analysis

Deep learning demonstrates impressive performance in many medical image analysis tasks. However, its reliability builds on the labeled medical datasets and the assumption of the same distributions between the training data (source domain) and the test data (target domain). Therefore, some unsupervised medical domain adaptation networks transfer knowledge from the source domain with rich labeled data to the target domain with only unlabeled data by learning domain-invariant features. We observe that conventional adversarial-training-based methods focus on the global distributions alignment and may overlook the class-level information, which will lead to negative transfer. In this paper, we attempt to learn the robust features alignment for the cross-domain medical image analysis. Specifically, in addition to a discriminator for alleviating the domain shift, we further introduce an auxiliary classifier to achieve robust features alignment with the class-level information. We first detect the unreliable target samples, which are far from the source distribution via diverse training between two classifiers. Next, a cross-classifier consistency regularization is proposed to align these unreliable samples and the negative transfer can be avoided. In addition, for fully exploiting the knowledge of unlabeled target data, we further propose a within-classifier consistency regularization to improve the robustness of the classifiers in the target domain, which enhances the unreliable target samples detection as well. We demonstrate that our proposed dual-consistency regularizations achieve state-of-the-art performance on multiple medical adaptation tasks in terms of both accuracy and Macro-F1-measure. Extensive ablation studies and visualization results are also presented to verify the effectiveness of each proposed module. For the skin adaptation results, our method outperforms the baseline and the second-best method by around 10 and 4 percentage points. Similarly, for the COVID-19 adaptation task, our model achieves consistently the best performance in terms of both accuracy (96.93%) and Macro-F1 (86.52%).

Time-frequency supervised contrastive learning via pseudo-labeling: An unsupervised domain adaptation network for rolling bearing fault diagnosis under time-varying speeds

The varying speed can cause the significant data distribution shift of bearings, making it difficult for deep learning-based bearing fault diagnosis models to ensure good generalization. Domain adaptation methods have been developed to address the domain shifts, while they struggle with the class-invariant features extraction under variable speed. Accordingly, a time–frequency supervised contrastive learning framework (TF-SupCon) is proposed for unsupervised cross-speed fault diagnosis of bearing. TF-SupCon adopts a pre-training-downstream task framework that aims to extract speed immune class-invariant features. During the pre-training phase, the physical consistency of the time-domain information and the frequency-domain information of bearing signal, a general feature applicable to different speed conditions, is learned in a targeted manner through supervised contrastive learning. Additionally, a K-Nearest Neighbor (KNN) algorithm based on cosine distance is designed to assign pseudo-labels to unlabeled target domain data, enabling cross-domain supervised contrastive pre-training. In the downstream task, unsupervised cross-domain fault diagnosis is performed using a KNN classifier and the speed immune time–frequency features by the trained encoders. It is worth noting that the same metric is maintained throughout both the pre-training phase and the downstream task, ensuring organic connection between the two stages. The superiority of TF-SupCon was demonstrated through a variety of fault diagnosis experiments conducted on both public and self-collected datasets. Lastly, the distances between time-domain and frequency-domain features of different categories were studied to verify the physical consistency between the features.

A Fault Detection Approach Based on One-Sided Domain Adaptation and Generative Adversarial Networks for Railway Door Systems.

Fault detection using the domain adaptation technique is one of the more promising methods of solving the domain shift problem, and has therefore been intensively investigated in recent years. However, the domain adaptation method still has elements of impracticality: firstly, domain-specific decision boundaries are not taken into consideration, which often results in poor performance near the class boundary; and secondly, information on the source domain needs to be exploited with priority over information on the target domain, as the source domain can provide a rich dataset. Thus, the real-world implementations of this approach are still scarce. In order to address these issues, a novel fault detection approach based on one-sided domain adaptation for real-world railway door systems is proposed. An anomaly detector created using label-rich source domain data is used to generate distinctive source latent features, and the target domain features are then aligned toward the source latent features in a one-sided way. The performance and sensitivity analyses show that the proposed method is more accurate than alternative methods, with an F1 score of 97.9%, and is the most robust against variation in the input features. The proposed method also bridges the gap between theoretical domain adaptation research and tangible industrial applications. Furthermore, the proposed approach can be applied to conventional railway components and various electro-mechanical actuators. This is because the motor current signals used in this study are primarily obtained from the controller or motor drive, which eliminates the need for extra sensors.

Application of transfer learning to predict drug-induced human in vivo gene expression changes using rat in vitro and in vivo data.

The liver is the primary site for the metabolism and detoxification of many compounds, including pharmaceuticals. Consequently, it is also the primary location for many adverse reactions. As the liver is not readily accessible for sampling in humans; rodent or cell line models are often used to evaluate potential toxic effects of a novel compound or candidate drug. However, relating the results of animal and in vitro studies to relevant clinical outcomes for the human in vivo situation still proves challenging. In this study, we incorporate principles of transfer learning within a deep artificial neural network allowing us to leverage the relative abundance of rat in vitro and in vivo exposure data from the Open TG-GATEs data set to train a model to predict the expected pattern of human in vivo gene expression following an exposure given measured human in vitro gene expression. We show that domain adaptation has been successfully achieved, with the rat and human in vitro data no longer being separable in the common latent space generated by the network. The network produces physiologically plausible predictions of human in vivo gene expression pattern following an exposure to a previously unseen compound. Moreover, we show the integration of the human in vitro data in the training of the domain adaptation network significantly improves the temporal accuracy of the predicted rat in vivo gene expression pattern following an exposure to a previously unseen compound. In this way, we demonstrate the improvements in prediction accuracy that can be achieved by combining data from distinct domains.

RUL prediction of rolling bearings across working conditions based on multi-scale convolutional parallel memory domain adaptation network

Rolling bearings are widely used in mechanical equipment, effectively determining the failure time of rolling bearings is particularly significant to ensure the safe performance of mechanical equipment. However, in industrial scenarios, the machine mainly works in the normal state for a long time, it is difficult to accumulate the same distribution of the whole life data, but the use of different distribution of data for forecasting will reduce the performance of deep learning-based prediction methods. Therefore, in order to tackle this problem, a multi-scale convolutional parallel memory domain adaptation network is investigated to forecast the residual useful life (RUL) of bearings across working conditions. Firstly, a new characteristic extractor—multi-scale convolutional parallel memory network is designed to extract spatial and temporal characteristics of bearing degradation data. At the same time, in order to minimize the distribution difference between source domain and target domain, a temporal-spatial feature alignment strategy is proposed to obtain domain invariable characteristics by combining maximum mean difference and domain adversarial learning. Finally, the availability of the proposed approach is verified using two rolling bearing data sets. The results reveal that it can efficiently forecast the RUL of rolling bearings across working conditions.

PDA: Progressive Domain Adaptation for Semantic Segmentation

The unsupervised domain adaptation semantic segmentation task is challenging due to the distribution shift problem between the source and the target domains. In this paper, we provide a novel perspective to address this problem. Currently, in the literature, it is shown that output-level domain adaptation networks (OL-DAN) can generate outputs with smaller distribution shift. Motivated by this phenomenon, a Progressive Domain Adaptation (PDA) framework is proposed, which uses the outputs generated by OL-DAN as the auxiliary domain images to progressively align the distribution between the source and target domains. Unlike existing two-stage input-level domain adaptation methods which use an image translation network as a standalone model to generate the auxiliary domain images, the PDA is an end-to-end framework that contains an OL-DAN and a domain fusion domain adaptation network (DF-DAN). The OL-DAN aims to gradually generate the outputs with smaller distribution shift and more accurate semantic structures as the auxiliary domain images in every iteration optimization. The DF-DAN is proposed to further mine the domain-invariant information from the auxiliary images and then fuse the features learned from the original images and the auxiliary images to obtain a richer representation. Finally, the distribution between the source and target domains is aligned to optimize the OL-DAN and DF-DAN. Experiments demonstrate that the proposed PDA achieves superior performance on three cross-domain semantic segmentation benchmarks.

Adversarial Distillation Adaptation Model with Sentiment Contrastive Learning for Zero-Shot Stance Detection

Zero-shot stance detection is both crucial and challenging because it demands detecting the stances of previously unseen targets in the inference stage. Learning transferable target invariant features effectively from training data is crucial for zero-shot stance detection. This paper proposes an adversarial adaptation approach for zero-shot stance detection, which applies an adversarial discriminative domain adaptation network to transfer knowledge efficiently. Specifically, the proposed model applies knowledge distillation to prevent overfitting the destination data and forgetting the learned source knowledge. Moreover, stance contrastive learning is applied to enhance the quality of feature representation for superior generalization, and sentiment information is extracted to assist with stance detection. The experimental results indicate that our model performs competitively on two benchmark datasets.

GLDAN: Global and Local Domain Adaptation Network for Cross-Wind Turbine Fault Diagnosis

Abstract Operating under harsh conditions and exposed to fluctuating loads for extended periods, wind turbines experience a heightened vulnerability in their key components. Early fault detection is crucial to enhance the reliability of wind turbines, minimize downtime, and optimize power generation efficiency. Although deep learning techniques have been widely applied to fault diagnosis tasks, yielding remarkable performance, practical implementations frequently confront the obstacle of acquiring a substantial quantity of labeled data to train an effective deep learning model. Consequently, this paper introduces an unsupervised global and local domain adaptation network (GLDAN) for fault diagnosis across wind turbines, enabling the model to efficiently transfer acquired knowledge to the target domain in the absence of labeled data. This feature renders it an appropriate solution for situations with limited labeled data availability. Employing adversarial training, GLDAN aligns global domain distributions, diminishing the overall discrepancy between source and target domains, and local domain distributions within a single fault category for both domains, capturing more intricate and specific fault features. The proposed approach is corroborated using actual wind farm data, and comprehensive experimental results demonstrate that GLDAN surpasses deep global domain adaptation methods in cross-wind turbine fault diagnosis, underlining its practical value in the field.

Multi-source information joint transfer diagnosis for rolling bearing with unknown faults via wavelet transform and an improved domain adaptation network

Recently, unsupervised domain adaptation fault diagnosis (FD) techniques, which learn transferable features by reducing distribution inconsistency of source and target domians, have gained abundant attention and greatly promoted the reliability of rolling bearing (RB) under variable operating conditions. However, open-set domain adaptation issues which contain unknown faults in the test set have not been well addressed. This paper presents a new semi-supervised FD method for RB by combining wavelet transform and an improved domain adaptation network. First, a multi-source domain adaptation network is proposed to extract rich transfer features and achieve complementary information from multiple sources. Then, a pseudo-margin vector is employed to handle unseen faults in the target domain and realize the accurate fault diagnosis of RB. Finally, a new loss function is designed by adding weights to the traditional maximum mean difference to make the common label set more compatible and combining a dynamic optimization strategy to adaptively update the loss of each part. Finally, two experiments indicate our proposed approach has a higher diagnosis accuracy and can effectively tackle the diagnosis issue of unseen faults across different working conditions.

A multi-constrained domain adaptation network for remaining useful life prediction of bearings

The prediction of the remaining useful life (RUL) of bearings could reduce maintenance costs and prevent serious consequences. Due to the variable working conditions of bearings, domain adaptation methods for cross-domain RUL prediction have received extensive attention. However, existing research often focuses solely on aligning marginal distributions while overlooking the alignment of conditional distributions and fails to retain domain-specific information during the alignment process. To address these issues, we propose a multi-constrained domain adaptation (MCDA) method for cross-domain RUL prediction. MCDA method is based on an adversarial domain adaptation architecture, where the improved domain discriminator aligns the data features of different degradation stages separately to achieve the alignment of marginal distribution and simultaneously aligns the conditional distribution. The state discrimination module is introduced to align the marginal and conditional distributions while retaining the target domain-specific information. To confirm the effectiveness of the proposed method, we conducted RUL prediction tasks for bearings in various cross-domain scenarios using both publicly and self-built datasets. The experimental results showed that the MCDA method could better handle cross-domain RUL prediction tasks compared to existing methods.