Adversarial Domain Adaptation Method Research Articles

Knowledge mapping-based adversarial domain adaptation: A novel fault diagnosis method with high generalizability under variable working conditions

Artificial intelligence-based fault diagnosis has recently been the subject of extensive research. However, the model learned from source data exhibits poor performance in target pattern recognition due to different data distributions caused by variable working conditions. Therefore, the transfer learning (TL) method, which reuses acquired knowledge and diagnoses the target domain fault without labels, has elicited the attention of researchers. The common deep TL method reduces the distance between the source and target domains in accordance with a certain divergence criterion that should be designed differently for specific tasks, leading to poor generalization results. In this study, we propose a knowledge mapping-based adversarial domain adaptation (KMADA) method with a discriminator and a feature extractor to generalize knowledge from target to source domain. The discriminator achieves the distance metric of the neural network wherein the target feature extractor maps the target data into the source feature space to explore domain-invariant knowledge. To accelerate the adversarial training process, KMADA fully utilizes the parameters obtained from the supervised pre-training. In addition, comparison analysis with other TL methods indicates the irreplaceable superiority of the KMADA, which achieves the highest diagnosis accuracy. Moreover, the visualization results demonstrate that the proposed model extracts the domain-invariant feature to realize knowledge mapping diagnosis, and thus, the model exhibits considerable research prospects.

Adversarial joint domain adaptation of asymmetric feature mapping based on least squares distance

Joint domain adaptation aims to learn a high-quality classifier for an unlabeled dataset with the help of auxiliary data. Most methods reduce domain shifts through some carefully designed distance measures. Adversarial learning, which is rarely used for joint domain adaptation, can learn more transferable features while avoiding explicit distance measures. However, it usually suffers from a gradient vanishing problem during the training process. In order to solve the above problems, we propose a novel adversarial joint domain adaptation method, namely Asymmetric Feature mapping based on Least Squares Distance (AFLSD), which consists of asymmetric marginal distribution alignment and conditional distribution alignment. The asymmetric feature mapping, which can get closer features with more flexible parameters, is optimized by the least squares distance to reduce the gradient vanishing problem. The results of classification and other comparative experiments show that AFLSD is superior to the most advanced domain adaptation methods.

A novel unsupervised adversarial domain adaptation network for remotely sensed scene classification

ABSTRACT High-resolution remote sensing scene classification is a widely applicable task. Due to the diversity of natural scenes and acquisition methods, in different satellite images, scenes of the same class are often variable in texture, background, illumination and spatial resolution. Thus, it is hard for a remote sensing sense database to contain enough representative images. In this case, the generalization error of traditional supervised classification methods might be too large to generate ideal results. Domain adaptation (DA) has been applied to image classification by reducing feature distribution discrepancy between the source domain (where labels are available) and the target domain (where images need to be classified). In this paper, we propose an unsupervised adversarial domain adaptation method boosted by a domain confusion network (ADA-BDC) which aims at adapting the images from different domains to appear as if drawn from the same domain and improve the transferability of our classifier. For this purpose, the feature extractor in ADA-BDC makes the source and target distributions closer by training a Generative Adversarial Nets (GAN) model. After that, a transferred classifier trained by transferred source domain features is able to acquire a better classification accuracy on the target domain than a non-transferred classifier. In this paper, the experiments are conducted on four remote sensing scene benchmark datasets which are different in spatial scale, resolution, land-cover pattern, etc. Experimental results show that our proposed method is able to improve the transferability across different datasets and improve the classification overall accuracy by 17.18% on average. The comparative experiments demonstrate that the proposed DA network outperforms the compared state-of-the-art domain adaptation methods on remote sensing image scene classification.

Unsupervised adversarial deep domain adaptation method for potato defects classification

In the last few years, deep learning methods have been proposed to automate the quality control of various agricultural products. Despite the excellent results obtained, one of their main drawbacks is the need for a large annotated dataset to obtain satisfactory performance. Building this dataset is time-consuming and tedious. Moreover, in some real-world applications, vision systems can be modified over time. Such changes may generate a drop in network performance trained with the initial database (source images). To avoid the creation of a new labeled database every time a change in data distribution occurs (target images), several domain adaptation methods have been proposed. In this article, we introduce an unsupervised deep domain adaptation method based on adversarial training. A large dataset is used, including six classes of potatoes: healthy, damaged, greening, black dot, common scab, and black scurf. Two different scenarios of domain adaptation problem are considered. Firstly, a simply modification of the image acquisition system is simulated by artificially increasing the brightness of some white potatoes images (target images). Secondly, a significantly different dataset including red potatoes is introduced. In this setting, white potatoes are used as source images and red tubers as target images. We propose to train a target classifier using a pseudo-label loss, due to the unavailability of target annotations. Experimental results show that a domain adaptation method is mandatory, going from an average F1-score of 0.46 without adaptation, to 0.84 by applying our method. Finally, a comparative analysis is achieved showing that adversarial-based unsupervised domain adaptation methods outperform discrepancy-based approaches.

WGAN domain adaptation for the joint optic disc-and-cup segmentation in fundus images.

The cup-to-disc ratio (CDR), a clinical metric of the relative size of the optic cup to the optic disc, is a key indicator of glaucoma, a chronic eye disease leading to loss of vision. CDR can be measured from fundus images through the segmentation of optic disc and optic cup . Deep convolutional networks have been proposed to achieve biomedical image segmentation with less time and more accuracy, but requires large amounts of annotated training data on a target domain, which is often unavailable. Unsupervised domain adaptation framework alleviates this problem through leveraging off-the-shelf labeled data from its relevant source domains, which is realized by learning domain invariant features and improving the generalization capabilities of the segmentation model. In this paper, we propose a WGAN domain adaptation framework for detecting optic disc-and-cup boundary in fundus images. Specifically, we build a novel adversarial domain adaptation framework that is guided by Wasserstein distance, therefore with better stability and convergence than typical adversarial methods. We finally evaluate our approach on publicly available datasets. Our experiments show that the proposed approach improves Intersection-over-Union score for optic disc-and-cup segmentation, Dice score and reduces the root-mean-square error of cup-to-disc ratio, when we compare it with direct transfer learning and other state-of-the-art adversarial domain adaptation methods. With this work, we demonstrate that WGAN guided domain adaptation obtains a state-of-the-art performance for the joint optic disc-and-cup segmentation in fundus images.

CLDA: an adversarial unsupervised domain adaptation method with classifier-level adaptation

CLDA: an adversarial unsupervised domain adaptation method with classifier-level adaptation

Feature transfer based adversarial domain adaptation method for cross-domain road extraction

In order to improve the cross-domain applicability of road segmentation, a feature transfer based adversarial domain adaptation method is presented for cross-domain road extraction. The presented method consists of two main parts, a feature transfer network and an adversarial domain adaption network. The feature transfer network transforms the images of the source domain into the feature space of the target domain. The adversarial domain adaption network learns to distinguish whether the results come from source domain or target domain, to promote the backbone network of road extraction to learn the common features of the road. The experimental results showed that the proposed method improved the generalization ability of the road extraction network and could extract the road target from cross-domain images accurately and effectively. The proposed method can realize the road extraction across domain without any annotation of target domain, so it has good application value.

Importance-weighted conditional adversarial network for unsupervised domain adaptation

In the construction of expert and intelligent systems, annotating and curating large datasets is very expensive; hence, there is a need to transfer the knowledge from existing annotated datasets to unlabeled data. However, data that are relevant for a specific application usually differ from publicly available datasets because they are sampled from a different domain. Domain adaptation (DA) has emerged as an efficient technique to compensate for such a domain shift. Recent studies have suggested that deep adversarial networks can achieve promising results for DA problems. However, existing adversarial DA methods assign equal importance to different examples and ignore the effect of difference in source domain samples or noise on adversarial performance. Moreover, most DA methods only focus on reducing the distribution difference, but not to learn a good target domain model. To address these issues, we propose an importance-weighted conditional adversarial (IWCA) network for unsupervised DA. In this study, an importance criterion based on domain similarity and prediction certainty is proposed to assign weights to different samples, which can reduce the harmful effects of difficult-to-transfer samples when reducing their cross-domain class conditional distribution differences. Furthermore, a sample selection criterion derived from the perspective of transfer cross validation is used to progressively select appropriate pseudo-labeled target samples to fine-tune the target model. These two criteria work in an EM-like manner that alternating align class conditional distribution for weighted samples and progressively select certain pseudo-labeled target samples to fine-tune the joint model. In this manner, the network will gradually generate features that approximate the actual conditional distribution of the target domain. The results of extensive experiments conducted on four datasets show that IWCA outperforms several state-of-the-art deep DA methods.

Spectral Normalization for Domain Adaptation

The transfer learning method is used to extend our existing model to more difficult scenarios, thereby accelerating the training process and improving learning performance. The conditional adversarial domain adaptation method proposed in 2018 is a particular type of transfer learning. It uses the domain discriminator to identify which images the extracted features belong to. The features are obtained from the feature extraction network. The stability of the domain discriminator directly affects the classification accuracy. Here, we propose a new algorithm to improve the predictive accuracy. First, we introduce the Lipschitz constraint condition into domain adaptation. If the constraint condition can be satisfied, the method will be stable. Second, we analyze how to make the gradient satisfy the condition, thereby deducing the modified gradient via the spectrum regularization method. The modified gradient is then used to update the parameter matrix. The proposed method is compared to the ResNet-50, deep adaptation network, domain adversarial neural network, joint adaptation network, and conditional domain adversarial network methods using the datasets that are found in Office-31, ImageCLEF-DA, and Office-Home. The simulations demonstrate that the proposed method has a better performance than other methods with respect to accuracy.

Triplet Adversarial Domain Adaptation for Pixel-Level Classification of VHR Remote Sensing Images

Pixel-level classification for very high resolution (VHR) images is a crucial but challenging task in remote sensing. However, since the diverse ways of satellite image acquisition and the distinct structures of various regions, the distributions of the same semantic classes among different data sets are dissimilar. Therefore, the classification model trained on one data set (source domain) may collapse, when it is directly applied to another one (target domain). To solve this problem, many adversarial-based domain adaptation methods have been proposed. However, these methods only consider the source and the target domains independently in the adversarial training, where only the target domain is explicitly contributed to narrow the gap between the distributions of both domains. Unlike previous methods, we propose a triplet adversarial domain adaptation (TriADA) method that jointly considers both domains to learn a domain-invariant classifier by a novel domain similarity discriminator. Specifically, the discriminator takes a triplet of segmentation maps as input, where two segmentation maps from the same domain are to be distinguished from the two maps from the different domains during the adversarial learning. Consequently, it explicitly considers both domains' information to narrow the distribution gap across domains. To enhance the discriminability of the classifier on the target domain, a class-aware self-training strategy, which depends on the output of the discriminator, is proposed to assign pseudo-labels with high adapted confidence on target data to retrain the classifier. Extensive experiments on several VHR pixel-level classification benchmarks demonstrate the effectiveness of our method as well as its superiority to the-state of the art.

Triplet Loss Guided Adversarial Domain Adaptation for Bearing Fault Diagnosis.

Recently, deep learning methods are becomingincreasingly popular in the field of fault diagnosis and achieve great success. However, since the rotation speeds and load conditions of rotating machines are subject to change during operations, the distribution of labeled training dataset for intelligent fault diagnosis model is different from the distribution of unlabeled testing dataset, where domain shift occurs. The performance of the fault diagnosis may significantly degrade due to this domain shift problem. Unsupervised domain adaptation has been proposed to alleviate this problem by aligning the distribution between labeled source domain and unlabeled target domain. In this paper, we propose triplet loss guided adversarial domain adaptation method (TLADA) for bearing fault diagnosis by jointly aligning the data-level and class-level distribution. Data-level alignment is achieved using Wasserstein distance-based adversarial approach, and the discrepancy of distributions in feature space is further minimized at class level by the triplet loss. Unlike other center loss-based class-level alignment approaches, which hasto compute the class centers for each class and minimize the distance of same class center from different domain, the proposed TLADA method concatenates 2 mini-batches from source and target domain into a single mini-batch and imposes triplet loss to the whole mini-batch ignoring the domains. Therefore, the overhead of updating the class center is eliminated. The effectiveness of the proposed method is validated on CWRU dataset and Paderborn dataset through extensive transfer fault diagnosis experiments.

Unsupervised Domain Adaptation With Adversarial Residual Transform Networks.

Domain adaptation (DA) is widely used in learning problems lacking labels. Recent studies show that deep adversarial DA models can make markable improvements in performance, which include symmetric and asymmetric architectures. However, the former has poor generalization ability, whereas the latter is very hard to train. In this article, we propose a novel adversarial DA method named adversarial residual transform networks (ARTNs) to improve the generalization ability, which directly transforms the source features into the space of target features. In this model, residual connections are used to share features and adversarial loss is reconstructed, thus making the model more generalized and easier to train. Moreover, a special regularization term is added to the loss function to alleviate a vanishing gradient problem, which enables its training process stable. A series of experiments based on Amazon review data set, digits data sets, and Office-31 image data sets are conducted to show that the proposed ARTN can be comparable with the methods of the state of the art.

Classifier-Constrained Deep Adversarial Domain Adaptation for Cross-Domain Semisupervised Classification in Remote Sensing Images

This letter presents a classifier-constrained deep adversarial domain adaptation (CDADA) method for cross-domain semisupervised classification in remote sensing (RS) images. A deep convolutional neural network (DCNN) is used to build feature representations to describe the semantic content of scenes before the adaptation process. Then, adversarial domain adaptation is used to align the feature distribution of the source and the target. Specifically, two different land-cover classifiers are used as a discriminator to consider land-cover decision boundaries between classes and increase their distance to separate them from the original land-cover class boundaries. The generator then creates robust transferable features far from the original land-cover class boundaries under the classifier constraint. The experimental results of six scenarios built from three benchmark RS scene data sets (AID, Merced, and RSI-CB data sets) are reported and discussed.

Adversarial domain adaptation for cross data source macromolecule in situ structural classification in cellular electron cryo-tomograms.

MotivationSince 2017, an increasing amount of attention has been paid to the supervised deep learning-based macromolecule in situ structural classification (i.e. subtomogram classification) in cellular electron cryo-tomography (CECT) due to the substantially higher scalability of deep learning. However, the success of such supervised approach relies heavily on the availability of large amounts of labeled training data. For CECT, creating valid training data from the same data source as prediction data is usually laborious and computationally intensive. It would be beneficial to have training data from a separate data source where the annotation is readily available or can be performed in a high-throughput fashion. However, the cross data source prediction is often biased due to the different image intensity distributions (a.k.a. domain shift).ResultsWe adapt a deep learning-based adversarial domain adaptation (3D-ADA) method to timely address the domain shift problem in CECT data analysis. 3D-ADA first uses a source domain feature extractor to extract discriminative features from the training data as the input to a classifier. Then it adversarially trains a target domain feature extractor to reduce the distribution differences of the extracted features between training and prediction data. As a result, the same classifier can be directly applied to the prediction data. We tested 3D-ADA on both experimental and realistically simulated subtomogram datasets under different imaging conditions. 3D-ADA stably improved the cross data source prediction, as well as outperformed two popular domain adaptation methods. Furthermore, we demonstrate that 3D-ADA can improve cross data source recovery of novel macromolecular structures.Availability and implementation https://github.com/xulabs/projects Supplementary information Supplementary data are available at Bioinformatics online.

Adversarial Domain Adaptation and Pseudo-Labeling for Cross-Modality Microscopy Image Quantification.

Cell or nucleus quantification has recently achieved state-of-the-art performance by using convolutional neural networks (CNNs). In general, training CNNs requires a large amount of annotated microscopy image data, which is prohibitively expensive or even impossible to obtain in some applications. Additionally, when applying a deep supervised model to new datasets, it is common to annotate individual cells in those target datasets for model re-training or fine-tuning, leading to low-throughput image analysis. In this paperSSS, we propose a novel adversarial domain adaptation method for cell/nucleus quantification across multimodality microscopy image data. Specifically, we learn a fully convolutional network detector with task-specific cycle-consistent adversarial learning, which conducts pixel-level adaptation between source and target domains and completes a cell/nucleus detection task. Then we generate pseudo-labels on target training data using the detector trained with adapted source images and further fine-tune the detector towards the target domain to boost the performance. We evaluate the proposed method on multiple cross-modality microscopy image datasets and obtain a significant improvement in cell/nucleus detection compared to the reference baselines and a recent state-of-the-art deep domain adaptation approach. In addition, our method is very competitive with the fully supervised models trained with all real target training labels.

基于深度对抗域适应的高分辨率遥感影像跨域分类

基于深度对抗域适应的高分辨率遥感影像跨域分类