Adversarial Transfer Learning Research Articles

Fault Diagnosis of Planetary Gearbox Based on Dynamic Simulation and Partial Transfer Learning.

To address the problem of insufficient real-world data on planetary gearboxes, which makes it difficult to diagnose faults using deep learning methods, it is possible to obtain sufficient simulation fault data through dynamic simulation models and then reduce the difference between simulation data and real data using transfer learning methods, thereby applying diagnostic knowledge from simulation data to real planetary gearboxes. However, the label space of real data may be a subset of the label space of simulation data. In this case, existing transfer learning methods are susceptible to interference from outlier label spaces in simulation data, resulting in mismatching. To address this issue, this paper introduces multiple domain classifiers and a weighted learning scheme on the basis of existing domain adversarial transfer learning methods to evaluate the transferability of simulation data and adaptively measure their contribution to label predictor and domain classifiers, filter the interference of unrelated categories of simulation data, and achieve accurate matching of real data. Finally, partial transfer experiments are conducted to verify the effectiveness of the proposed method, and the experimental results show that the diagnostic accuracy of this method is higher than existing transfer learning methods.

Multi-source adversarial transfer learning for ultrasound image segmentation with limited similarity

Lesion segmentation of ultrasound medical images based on deep learning techniques is a widely used method for diagnosing diseases. Although there is a large amount of ultrasound image data in medical centers and other places, labeled ultrasound datasets are a scarce resource, and it is likely that no datasets are available for new tissues/organs. Transfer learning provides the possibility to solve this problem, but there are too many features in natural images that are not related to the target domain. As a source domain, redundant features that are not conducive to the task will be extracted. Migration between ultrasound images can avoid this problem, but there are few types of public datasets, and it is difficult to find sufficiently similar source domains. Compared with natural images, ultrasound images have less information, and there are fewer transferable features between different ultrasound images, which may cause negative transfer. To this end, a multi-source adversarial transfer learning network for ultrasound image segmentation is proposed. Specifically, to address the lack of annotations, the idea of adversarial transfer learning is used to adaptively extract common features between a certain pair of source and target domains, which provides the possibility to utilize unlabeled ultrasound data. To alleviate the lack of knowledge in a single source domain, multi-source transfer learning is adopted to fuse knowledge from multiple source domains. In order to ensure the effectiveness of the fusion and maximize the use of precious data, a multi-source domain independent strategy is also proposed to improve the estimation of the target domain data distribution, which further increases the learning ability of the multi-source adversarial migration learning network in multiple domains. The effectiveness of multi-source adversarial transfer learning is demonstrated through experiments on three datasets of ultrasound image datasets.

Quantum Adversarial Transfer Learning.

Adversarial transfer learning is a machine learning method that employs an adversarial training process to learn the datasets of different domains. Recently, this method has attracted attention because it can efficiently decouple the requirements of tasks from insufficient target data. In this study, we introduce the notion of quantum adversarial transfer learning, where data are completely encoded by quantum states. A measurement-based judgment of the data label and a quantum subroutine to compute the gradients are discussed in detail. We also prove that our proposal has an exponential advantage over its classical counterparts in terms of computing resources such as the gate number of the circuits and the size of the storage required for the generated data. Finally, numerical experiments demonstrate that our model can be successfully trained, achieving high accuracy on certain datasets.

Smart contracts vulnerability detection model based on adversarial multi-task learning

Vulnerability detection is important for smart contracts because of their immutable and irreversible features. In this work, a new detection method based on adversarial multi-task learning is proposed to improve the accuracy of existing vulnerability detection methods, which is based on the multi-task learning framework, including a shared part and a task-specific part. We optimize the multi-task learning frameworks and propose the mixed parameter sharing method to make each task not only maintain its uniqueness, but also share features with other tasks, which helps solve the problem that the hard parameter sharing method cannot constrain the underlying shared layer and improve the quality of extracted features. In addition, we introduce adversarial transfer learning to reduce noise pollution caused by the private feature and interference between the general feature and the private feature. We experimented on datasets obtained from our previous work, and the experimental results prove that our proposed model can judge whether there are vulnerabilities in smart contracts and then identify their types. Additionally, the results also show that our model effectively improves detection accuracy and has an advantage in performance over representative methods.

Acoustic Sensor Placement Optimization for Compressor Based on Adversarial Transfer Learning and Vibro-Acoustic Simulation

Generally, the success of optimal sensor placement (OSP) method based on artificial intelligence highly depends on signals at all feasible placements, which may be unavailable or expensive. Therefore, the acoustic sensor placement optimization method is proposed based on adversarial transfer learning (ATL) and vibro-acoustic simulation. First, the vibro-acoustic simulation is applied to provide sufficient simulation signals for all feasible placements. To bridge the deviation of simulation and measured signals, the ATL-based conditional generative adversarial network (CGAN) is presented, which can transfer signals from limited measured placements to unmeasured placements. In addition, a multi objective optimization model is proposed to obtain the OSP from three aspects, and it is useful for structural health monitoring (SHM). The acoustic signals obtained from the experimental platform are utilized to explore the feasibility and effectiveness of the proposed method. It can accurately detect the fault with average accuracy of 98.35% under four working conditions. The comparison investigations demonstrate that the proposed method can obtain high-quality signals at all feasible placements, which can realize SHM with the optimal sensor placements and the least sensor cost.

Geo‐environment‐aware adversarial transfer learning method for landslide susceptibility evaluation of complex mountainous areas

AbstractLandslide susceptibility evaluation (LSE) is a critical issue for disaster prevention. Limited by labor cost and observation technology, landslide samples are extremely limited in dense vegetation‐covered and remote areas, making the common supervised learning model underfit with limited samples. Therefore, the reliability of analysis results in mountainous areas is low. Transfer learning can achieve reliable assessment without the need for representative samples. However, transfer learning suffers from environmental heterogeneity in regional LSE and may transfer incorrect classification knowledge of landslide features from dissimilar environments. Aiming at these challenges, we proposed a geo‐environment‐aware LSE method based on unsupervised adversarial transfer learning. The key is to consider the difference in landslide features in different geo‐environments. The study areas were first divided into multiple sub‐environments, and the similarity between the sub‐environments was calculated. Then an environment‐aware adversarial transfer model was built for fine‐grained aligning of the landslide feature with similar sub‐environments and for reducing negative transfer between dissimilar environments. The fitted classification model was employed to predict the target regions and to generate the final LSE. The experimental results indicated that the proposed method achieves reliable LSE for sample‐free regions. The accuracy of the proposed method is 7–12% better than commonly used methods such as support vector machines, random forests, and artificial neural networks. The performance of the proposed method is even close to the results of supervised learning with the presence of representative samples, and it also performs more globally and objectively in susceptibility mapping. These results reveal that the proposed method effectively transfers the knowledge of landslide susceptibility from other regions to the sample‐free region.

Digital twin aided adversarial transfer learning method for domain adaptation fault diagnosis

Machine health management has become the focus of equipment monitoring upgrading with the advance of digital twin (DT). The DT model is able to generate system performance data that is close to reality, which opens a new way for the cyber-physical integration of equipment monitoring. Furthermore, it also provides a significant opportunity for mechanical fault diagnosis when the collected fault signals are insufficient. In this paper, a DT aided intelligent fault diagnosis model is proposed for triplex pump. Specifically, the simulation model of the triplex pump is built by Simscape in MATLAB, and the measured simulation data is continuously updated to construct the DT model. Then a novel transfer learning model based on domain-adversarial strategy and Wasserstein distance is present and trained by the source domain data which generated from the DT model. Next, the opening pressure of the triplex pump is controlled to simulate different working conditions, so as to achieve feature transfer and fault diagnosis for the DT model. The experimental results show that the proposed method is effective and superior to other advanced transfer learning methods.

Adversarial Transferred Data-Assisted Soft Sensor for Enhanced Multigrade Quality Prediction.

Although recent transfer learning soft sensors show promising applications in multigrade chemical processes, good prediction performance mainly relies on available target domain data, which is difficult to achieve for a start-up grade. Additionally, only employing a single global model is inadequate to characterize the inner relationship of process variables. A just-in-time adversarial transfer learning (JATL) soft sensing method is developed to enhance multigrade process prediction performance. The distribution discrepancies of process variables between two different operating grades are first reduced by the ATL strategy. Subsequently, by applying the just-in-time learning approach, a similar data set is selected from the transferred source data for reliable model construction. Consequently, with the JATL-based soft sensor, quality prediction of a new target grade is implemented without its own labeled data. Experimental results on two multigrade chemical processes validate that the JATL method can give rise to the improvement of model performance.

Conditional distribution-guided adversarial transfer learning network with multi-source domains for rolling bearing fault diagnosis

The application of transfer learning to effectively identify rolling bearing fault has been attracting much attention. Most of the current studies are based on single-source domain or multi-source domains constructed from different working conditions of the same machine. However, in practical scenarios, it is common to obtain multiple source domains from different machines, which brings new challenges to how to use these source domains to complete fault diagnosis. To solve the issue, a conditional distribution-guided adversarial transfer learning network with multi-source domains (CDGATLN) is developed for fault diagnosis of bearing installed on different machines. Firstly, the knowledge of multi-source domains from different machines is transferred to the single target domain by decreasing data distribution discrepancy between each source domain and target domain. Then, a conditional distribution-guided alignment strategy is introduced to decrease conditional distribution discrepancy and calculate the importance per source domain based on the conditional distribution discrepancy, so as to promote the knowledge transfer of each source domain. Finally, a monotone importance specification mechanism is constructed to constrain each importance to ensure that the source domain with low importance will not be discarded, which enables the knowledge of each source domain to participate in the construction of the model. Extensive experimental results verify the effectiveness and superiority of CDGATLN.

Cross Domain Hybrid Feature Fusion based Sarcastic Opinion Recognition Over E-Commerce Reviews Using Adversarial Transfer Learning

Cross Domain Hybrid Feature Fusion based Sarcastic Opinion Recognition Over E-Commerce Reviews Using Adversarial Transfer Learning

Research on Named Entity Recognition for Spoken Language Understanding Using Adversarial Transfer Learning

In this paper, we propose an adversarial transfer learning method to solve the lack of data resources for named entity recognition (NER) tasks in spoken language understanding. In the framework, we use bi-directional long short-term memory with self-attention and conditional random field (BiLSTM-Attention-CRF) model which combines character and word information as the baseline model to train source domain and target domain corpus jointly. Shared features between domains are extracted by a shared feature extractor. This paper uses two different sharing patterns simultaneously: full sharing mode and private sharing mode. On this basis, an adversarial discriminator is added to the shared feature extractor to simulate generative adversarial networks (GAN) and eliminate domain-dependent features. This paper compares ordinary adversarial discriminator (OAD) and generalized resource-adversarial discriminator (GRAD) through experiments. The experimental results show that the transfer effect of GRAD is better than other methods. The F1 score reaches 92.99% at the highest, with a relative increase of 12.89%. It can effectively improve the performance of NER tasks in resource shortage fields and solve the problem of negative transfer.

AMTLDC: a new adversarial multi-source transfer learning framework to diagnosis of COVID-19.

In recent years, deep learning techniques have been widely used to diagnose diseases. However, in some tasks, such as the diagnosis of COVID-19 disease, due to insufficient data, the model is not properly trained and as a result, the generalizability of the model decreases. For example, if the model is trained on a CT scan dataset and tested on another CT scan dataset, it predicts near-random results. To address this, data from several different sources can be combined using transfer learning, taking into account the intrinsic and natural differences in existing datasets obtained with different medical imaging tools and approaches. In this paper, to improve the transfer learning technique and better generalizability between multiple data sources, we propose a multi-source adversarial transfer learning model, namely AMTLDC. In AMTLDC, representations are learned that are similar among the sources. In other words, extracted representations are general and not dependent on the particular dataset domain. We apply the AMTLDC to predict Covid-19 from medical images using a convolutional neural network. We show that accuracy can be improved using the AMTLDC framework, and surpass the results of current successful transfer learning approaches. In particular, we show that the AMTLDC works well when using different dataset domains, or when there is insufficient data.

Development of an Adversarial Transfer Learning-Based Soft Sensor in Industrial Systems

Development of an Adversarial Transfer Learning-Based Soft Sensor in Industrial Systems

Spectrum sensing based on adversarial transfer learning

Spectrum sensing based on adversarial transfer learning

An underwater dam crack image segmentation method based on multi-level adversarial transfer learning

Crack detection is necessary to ensure the health of dams. Traditional detection methods perform poorly because of weak adaptability and poor image quality. Deep learning shows excellent performance in crack image detection. However, it is difficult to realize supervised learning due to the lack of labelled underwater crack image datasets. Thus, a transfer learning method named MA-AttUNet is proposed. The proposed method realizes knowledge transfer of crack image features using a multi-level adversarial transfer network. With this method, prior knowledge learned from the source domain can be applied to underwater crack image segmentation. Additionally, the attention mechanism is integrated into the segmentation network to eliminate noise interference during detection by assigning different weights to target and background pixels. Experiments show that the proposed method achieves higher segmentation precision than existing works.

Generic semi-supervised adversarial subject translation for sensor-based activity recognition

Performance of Human Activity Recognition (HAR) models, particularly deep neural networks, is highly contingent upon the availability of the massive amount of annotated training data. Though, data collection and manual labeling in the HAR domain are prohibitively expensive due to human resource dependence in both steps. Hence, domain adaptation techniques are proposed to adapt the knowledge from the existing source of data. More recently, adversarial transfer learning methods have shown promising results for visual classification, yet limited for HAR problems, which are still prone to the unfavorable effects of the imbalanced distribution of samples. This paper presents a novel generic semi-supervised approach that takes advantage of the adversarial framework to tackle these shortcomings by leveraging knowledge from annotated samples exclusively from the source subject and unlabeled ones of the target subject. An extensive subject translation experiments is conducted on three large, middle, and small-size datasets with different levels of imbalance to assess the robustness of the proposed model to the scale as well as the imbalance in the data. The results demonstrate the effectiveness of our proposed algorithms over state-of-the-art methods, which led to up to 13%, 4%, and 13% improvement of our high-level activities recognition metrics for Opportunity, LISSI, and PAMAP2 datasets, respectively.

An ensemble and shared selective adversarial network for partial domain fault diagnosis of machinery

Recently, some adversarial transfer learning (TL) approaches have been developed for addressing the partial domain adaptation (DA) problems in machinery fault diagnosis. However, these existing methods generally follow the partial DA framework of multiple sub-domain discriminators, which causes the overly complex model in dealing with many source classes. Moreover, the classifier mostly based on a single intelligent model with limited generalization, it may predict the wrong class-probability for unlabeled samples, which will cause the negative transfer when used in the model training. In response to the challenges, an ensemble and shared selective adversarial network (ES-SAN) is proposed in this paper. In this network, by introducing a correlation layer to correlate both class and domain informations for each sample, a single-intelligent model based shared module is constructed, which can transform between the classifier and the discriminator capable of multi sub-domain discrimination, so as to form a simplified partial DA framework. Moreover, multiple shared modules based on different intelligent models are integrated into an ensemble module, which can output reliable probability weights to promote positive transfer. Experimental investigations on two diagnosis datasets demonstrate that the proposed ES-SAN outperforms the existing methods in the partial DA diagnosis.

A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information.

Cancer prognosis is an essential goal for early diagnosis, biomarker selection, and medical therapy. In the past decade, deep learning has successfully solved a variety of biomedical problems. However, due to the high dimensional limitation of human cancer transcriptome data and the small number of training samples, there is still no mature deep learning-based survival analysis model that can completely solve problems in the training process like overfitting and accurate prognosis. Given these problems, we introduced a novel framework called SAVAE-Cox for survival analysis of high-dimensional transcriptome data. This model adopts a novel attention mechanism and takes full advantage of the adversarial transfer learning strategy. We trained the model on 16 types of TCGA cancer RNA-seq data sets. Experiments show that our module outperformed state-of-the-art survival analysis models such as the Cox proportional hazard model (Cox-ph), Cox-lasso, Cox-ridge, Cox-nnet, and VAECox on the concordance index. In addition, we carry out some feature analysis experiments. Based on the experimental results, we concluded that our model is helpful for revealing cancer-related genes and biological functions.

A simulation-data-driven subdomain adaptation adversarial transfer learning network for rolling element bearing fault diagnosis

In recent years, increasing numbers of deep learning methods for fault diagnosis of rolling element bearings (REBS) have been proposed. However, in industry, the scarcity of available data to monitor the health condition of REBS leads to low recognition accuracy of the trained intelligent diagnostic models. To solve this problem, we propose a simulation-data-driven subdomain adaptation adversarial transfer learning (TL) network (SAATLN). Firstly, a defect vibration model is introduced to simulate vibration signals of different types of REBS faults. And the real signal and simulated signal are used as the target domain and source domain of the TL fault diagnosis methods, respectively. Secondly, SAATLN uses the designed residual squeeze-and-excitation (Re-SE) blocks to extract transfer features between different domains. Meanwhile, it combines adversarial learning and subdomain adaptation to adapt the marginal distribution and conditional distribution discrepancies of high-level features. Also, the local maximum mean discrepancy is introduced as the subdomain adaptation metric criterion. Finally, different transfer tasks are performed on the artificially damaged and run-to-failure REBS data sets. The results demonstrate the effectiveness and superiority of the SAATLN in the simulation-data-driven REBS fault diagnosis.

Self-supervised bi-classifier adversarial transfer network for cross-domain fault diagnosis of rotating machinery

In real industrial scenarios, deep learning-based fault diagnosis has been a popular topic lately. Unfortunately, the source-trained model typically usually underperforms in target domain owning to changeable working conditions. To resolve this problem, a novel self-supervised bi-classifier adversarial transfer learning (SBATL) network by introducing self-supervised learning (SSL) and class-conditional entropy minimization is presented. Concretely, the SBATL is made up of a feature extractor, a discrepancy detector of two classifiers, and a clustering metric based on SSL, which jointly conducts self-supervised and supervised optimization in a two-stream training procedure. In the self-supervised stream, target pseudo labels obtained by SSL are used to construct the topological clustering metric for target feature optimization. In the supervised stream, the feature extractor and classifiers compete with each other in adversarial training, which bridges the discrepancy between two classifiers. Additionally, the class-conditional entropy minimization of target domain is further embedded into both streams to amend the decision boundaries of two classifiers to pass low-density regions. The results indicate that the SBATL gets better cross-domain fault diagnosis performances when compared with other popular methods.