Open Set Domain Adaptation Research Articles

Dual teacher–student based separation mechanism for open set domain adaptation

Open set domain adaptation (OSDA) is a more realistic and challenging domain adaptation protocol that assumes that the label spaces of source and target domains are different. Existing OSDA methods lack the ability to identify unknown samples and suffer more from negative transfer. To mitigate the negative transfer, we propose a dual teacher–student based sample separation mechanism to accurately identify and separate known and unknown samples in the target domain. Specifically, in the teacher–student structure (TS), students are insensitive to target unknown samples because they learn only source domain knowledge. Therefore, we exploit this property of students to construct a dual teacher–student structure (DTS), which quantifies the similarity between the target samples and the source domain and improves the accuracy and robustness of the similarity assessment. In addition, we further build a selective adversarial learning model by weighting the target samples with the weights derived from the similarity assessment. This promotes cross-domain alignment of the distribution of target samples with higher weights and the distribution of source samples. Extensive experiments conducted on the benchmark dataset demonstrate the significant advantages of the proposed approach over the extant methods.

TESPOSDA-SEI: tensor embedding substructure preserving open set domain adaptation for specific emitter identification

TESPOSDA-SEI: tensor embedding substructure preserving open set domain adaptation for specific emitter identification

Statistical identification guided open-set domain adaptation in fault diagnosis

Statistical identification guided open-set domain adaptation in fault diagnosis

Open Set Domain Adaptation With Soft Unknown-Class Rejection.

The goal of domain adaptation (DA) is to train a good model for a target domain, with a large amount of labeled data in a source domain but only limited labeled data in the target domain. Conventional closed set domain adaptation (CSDA) assumes source and target label spaces are the same. However, this is not quite practical in real-world applications. In this work, we study the problem of open set domain adaptation (OSDA), which only requires the target label space to partially overlap with the source label space. Consequently, the solution to OSDA requires unknown classes detection and separation, which is normally achieved by introducing a threshold for the prediction of target unknown classes; however, the performance can be quite sensitive to that threshold. In this article, we tackle the above issues by proposing a novel OSDA method to perform soft rejection of unknown target classes and simultaneously match the source and target domains. Extensive experiments on three standard datasets validate the effectiveness of the proposed method over the state-of-the-art competitors.

A Multi-Source Weighted Deep Transfer Network for Open-Set Fault Diagnosis of Rotary Machinery

In real industries, there often exist application scenarios where the target domain holds fault categories never observed in the source domain, which is an open-set domain adaptation (DA) diagnosis issue. Existing DA diagnosis methods under the assumption of sharing identical label space across domains fail to work. What is more, labeled samples can be collected from different sources, where multisource information fusion is rarely considered. To handle this issue, a multisource open-set DA diagnosis approach is developed. Specifically, multisource domain data of different operation conditions sharing partial classes are adopted to take advantage of fault information. Then, an open-set DA network is constructed to mitigate the domain gap across domains. Finally, a weighting learning strategy is introduced to adaptively weigh the importance on feature distribution alignment between known class and unknown class samples. Extensive experiments suggest that the proposed approach can substantially boost the performance of open-set diagnosis issues and outperform existing diagnosis approaches.

Open set domain adaptation with latent structure discovery and kernelized classifier learning

Numerous visual domain adaptation methods have been proposed for transferring knowledge from a well-labeled source domain to an unlabeled but related target domain. Most of existing works are only geared to closed set domain adaptation, where an identical label space is shared between two domains. In this paper, we focus on a more realistic but challenging scenario, open set domain adaptation, where the target domain contains unknown classes that do not appear in the label space of source domain. The main task of open set domain adaptation is to simultaneously recognize the target images of known classes and those of unknown classes correctly. To achieve this goal, in this paper, we propose a novel open set domain adaptation method, which consists of two parts: latent structure discovery and kernelized classifier learning. In the first part, we employ an adaptive discriminative graph learning strategy to capture the intrinsic manifold structure of the source and target domain data in the latent feature space, such that the boundaries among all classes will be delineated more clearly. In the second part, the samples from the latent feature space are mapped into a high-dimensional kernel space to make them linearly separable, and a linear classifier is learned by jointly operating unknown target samples separating, known samples matching and local structure preserving. As the optimization problem is not convex with all variables, we devise an efficient iterative algorithm to solve it. The extensive experimental results on five image datasets confirm the superiority of the proposed method compared with the state-of-the-art traditional and deep competitors.

Sample separation and domain alignment complementary learning mechanism for open set domain adaptation

Sample separation and domain alignment complementary learning mechanism for open set domain adaptation

Deep learning-based open set domain adaptation for plant disease recognition across multiple greenhouse scenarios

Deep learning-based open set domain adaptation for plant disease recognition across multiple greenhouse scenarios

Open Set Domain Adaptation via Instance Affinity Metric and Fine-Grained Alignment for Remote Sensing Scene Classification

In the practical application of remote sensing scene classification (RSSC), domain adaptation is introduced to handle the situation where the distribution of training data (source) and test data (target) is different. Compared to general domain adaptation, open set domain adaptation (OSDA) is suitable for more realistic situations where there are additional unknown classes in the target domain. The key to solving this problem is to separate unknown samples from target data to avoid negative transfer caused by mismatching unknown/known samples. In this letter, we propose a novel and effective method, named Instance Affinity metric-based Fine-grained Adaptation Network (IAFAN), for OSDA in RSSC. Concretely, an Unknown Sample Separation (USS) mechanism based on the instance affinity-aware matrix is pioneeringly proposed to endow the model with the ability to distinguish unknown samples. In addition, considering the high inter-class similarity and rich intra-class diversity of remote sensing images, we introduce the Sample Discriminability Enhancement (SDE) loss to further increase the inter-class distance and narrow the intra-class difference, thereby alleviating the negative transfer caused by sample misclassification and mismatching. In the feature confusion stage, we specially design Mask-mmd for OSDA as an adaptation metric to conduct semantic fine-grained cross-domain alignment of known samples while keeping unknown samples out of alignment, which avoids negative transfer during the adaptation process. Finally, we evaluate IAFAN on transfer tasks between different public remote sensing datasets, and the results verify that our method significantly outperforms previous methods in RSSC.

Manifold Regularized Joint Transfer for Open Set Domain Adaptation

Unsupervised Domain Adaptation (UDA) aims to leverage knowledge of a well-labeled source domain to learn an effective classifier for an unlabeled target domain. However, a common scenario in real-world applications is that the target domain contains unknown categories that are not observed in the source domain. This setting is termed as open set domain adaptation (OSDA). Most existing approaches of OSDA can only classify known classes well but fail to recognize unknown samples effectively. In this paper, we propose an effective method, named manifold regularized joint transfer (MRJT), for OSDA. MRJT learns new feature representations by simultaneously reducing distribution discrepancy between domains, increasing compactness of within-class, discriminating different known classes, and distinguishing the unknown from the known. The learned new features are projected onto reproducing kernel Hilbert space. In this space, a weighted structural risk minimization method is integrated with manifold regularization to utilize geometric information sufficiently to learn an effective classifier. Extensive experimental results on four real-world datasets verify the superiority of our method. It can not only classify known samples into the right known classes but also recognize unknown samples effectively.

Semantic Consistent Embedding for Domain Adaptive Zero-Shot Learning.

Unsupervised domain adaptation has limitations when encountering label discrepancy between the source and target domains. While open-set domain adaptation approaches can address situations when the target domain has additional categories, these methods can only detect them but not further classify them. In this paper, we focus on a more challenging setting dubbed Domain Adaptive Zero-Shot Learning (DAZSL), which uses semantic embeddings of class tags as the bridge between seen and unseen classes to learn the classifier for recognizing all categories in the target domain when only the supervision of seen categories in the source domain is available. The main challenge of DAZSL is to perform knowledge transfer across categories and domain styles simultaneously. To this end, we propose a novel end-to-end learning mechanism dubbed Three-way Semantic Consistent Embedding (TSCE) to embed the source domain, target domain, and semantic space into a shared space. Specifically, TSCE learns domain-irrelevant categorical prototypes from the semantic embedding of class tags and uses them as the pivots of the shared space. The source domain features are aligned with the prototypes via their supervised information. On the other hand, the mutual information maximization mechanism is introduced to push the target domain features and prototypes towards each other. By this way, our approach can align domain differences between source and target images, as well as promote knowledge transfer towards unseen classes. Moreover, as there is no supervision in the target domain, the shared space may suffer from the catastrophic forgetting problem. Hence, we further propose a ranking-based embedding alignment mechanism to maintain the consistency between the semantic space and the shared space. Experimental results on both I2AwA and I2WebV clearly validate the effectiveness of our method.

Heterogeneous domain adaptation for intracortical signal classification using domain consensus

Most domain adaptation studies have made efforts to solve the problem of homogeneous data shifts, which share the same feature and label space in both domains. However, for the heterogeneous scenarios, it still needs improvements in practical applications for intracortical brain-computer interface (iBCI) systems. To address this issue, we propose a domain adaptation framework for decoding reaching and grasping movements of iBCI systems in two heterogeneous scenarios, i.e., partial domain adaptation (PDA) and open set domain adaptation (OSDA). The proposed framework combined the neural activity vector (NAV) features and the domain consensus clustering (DCC) method for iBCI classification. For heterogeneous scenarios, the framework can map common classes according to the clusters with the highest domain consensus score at the semantic-level and distinguish private classes at the sample-level. We evaluated the framework on the data, which were collected from three array electrodes implanted in three regions of cerebral cortexes from one monkey across two sessions, each consisting of four continuous days. The proposed framework outperformed than four PDA and two OSDA methods, which effectively reduced the disparity of feature and label space in two heterogeneous scenarios. Consequently, our results demonstrated that the proposed framework is a new solution for decoder calibration in two heterogeneous scenarios, and it facilitates the generality of the iBCI system for clinical applications.

Extending Partial Domain Adaptation Algorithms to the Open-Set Setting

Partial domain adaptation (PDA) is a framework for mitigating the covariate shift problem when target labels are contained in source labels. For this task, adversarial neural network (ANN) methods proposed in the literature have been proven to be flexible and effective. In this work, we adapt such methods to tackle the more general problem of open-set domain adaptation (OSDA), which further allows the existence of target instances with labels outside the source labels. The aim in OSDA is to mitigate the covariate shift problem and to identify target instances with labels outside the source label space. We show that the effectiveness of ANN methods utilized in the PDA setting is hindered by outlier target instances, and we propose an adaptation for effective OSDA.

GCL-OSDA: Uncertainty prediction-based graph collaborative learning for open-set domain adaptation

Open-set domain adaptation (OSDA), which allows the target domain to store invisible class samples in the source domain, has recently received significant attention. In this paper, we propose a new unsupervised OSDA classification framework using an evidential network and multi-binary classifier and consider their jointly selected samples as a pseudo-labelled sample set of an unknown class. Specifically, this study designed an evidential network based on the D-S evidence theory to predict the degree of belief that a sample belongs to an unknown class. By selecting samples with high-uncertainty, false positive samples can be removed, which improves the reliability of unknown sample selection. Then, to better explore the intra-class relationship, an open-set graph convolutional network (OSGC) is proposed to extract distinguishable features of known and unknown samples in a weighted adversarial adaptation manner. Moreover, this paper presents a graph collaborative learning strategy to retrain the unknown recognition module (URM) with high confidence pseudo-labelled samples, which is predicted by the graph convolution network (GCN), where the target known class distribution is learned. Experimental results show that the proposed method outperforms state-of-the-art OSDA algorithms on three benchmark datasets and maintains a high recognition accuracy for unknown classes over a wide range of openness.

Adversarial Fuzzy-Weighted Deep Transfer Learning for Intelligent Damage Diagnosis of Bridge With Multiple New Damages

Recently, domain-adaptation based transfer learning has been extensively studied and successfully achieved promising results in addressing the domain drift in closed-set scenarios. However, in the bridge damage diagnosis field, the target data-sets collected from bridges frequently present samples of new damages that were not observed in the source domain, which is known as the open-set domain adaptation problem. To address this problem, this paper proposes a new open-set deep transfer learning algorithm based on joint weighted sub-domain adaptation. First, a joint weighting mechanism is proposed based on adversarial learning and fuzzy theory to represent the similarity of target-domain samples with source-domain classes, and explore the method of separating the known and unknown classes in the target domain to solve the negative transfer problem. Then, to capture the fine-grained transferable information, a sub-domain adaptation algorithm based on minimizing the multi-channel multi-kernel weighted local maximum mean discrepancy (MCMK-WLMMD) is proposed to align the corresponding sub-domains in the two domains. Finally, membership is introduced to build an unsupervised fuzzy clustering model with evaluation indicator to recognize multiple unknown damages. Extensive experiments on open-set transfer tasks between three bridges verify the effectiveness of the algorithm.

Open-set domain adaptation by deconfounding domain gaps

Open-Set Domain Adaptation (OSDA) aims to adapt the model trained on a source domain to the recognition tasks in a target domain while shielding any distractions caused by open-set classes, i.e., the classes “unknown” to the source model. Compared to standard DA, the key of OSDA lies in the separation between known and unknown classes. Existing OSDA methods often fail the separation because of overlooking the confounders (i.e., the domain gaps), which means their recognition of “unknown classes” is not because of class semantics but domain difference (e.g., styles and contexts). We address this issue by explicitly deconfounding domain gaps (DDP) during class separation and domain adaptation in OSDA. The mechanism of DDP is to transfer domain-related styles and contexts from the target domain to the source domain. It enables the model to recognize a class as known (or unknown) because of the class semantics rather than the confusion caused by spurious styles or contexts. In addition, we propose a module of ensembling multiple transformations (EMT) to produce calibrated recognition scores, i.e., reliable normality scores, for the samples in the target domain. Extensive experiments on two standard benchmarks verify that our proposed method outperforms a wide range of OSDA methods, because of its advanced ability of correctly recognizing unknown classes.

Feature distance-based deep prototype network for few-shot fault diagnosis under open-set domain adaptation scenario

Transfer learning-based fault diagnosis methods have revealed prominent application prospects. However, most of these methods can achieve efficient knowledge transfer across domains only when the source and target domains share the exact identical label space. In practical applications, the model trained on the source domain may be faced with a larger target domain, because unknown fault types may emerge unpredictably in testing stage. Besides, a large amount of labeled data is tough to obtain for training. To address these problems, a feature distance-based deep prototype network (FDDPN) is proposed for few-shot fault diagnosis under open-set domain adaptation scenario. The proposed method contains three stages for training. The prototype of each known class is calculated via novel k-means feature clustering algorithm in the first stage. In the second stage, the few labeled target samples are assigned to the class of the nearest prototype according to the feature distance. Finally, compactness between the few target labeled samples and the corresponding prototypes is enhanced in the third stage. Experimental results demonstrated that the proposed method can realize accurate classification of known classes and effective rejection of unknown classes.

Self-Labeling Framework for Novel Category Discovery over Domains

Unsupervised domain adaptation (UDA) has been highly successful in transferring knowledge acquired from a label-rich source domain to a label-scarce target domain. Open-set domain adaptation (open-set DA) and universal domain adaptation (UniDA) have been proposed as solutions to the problem concerning the presence of additional novel categories in the target domain. Existing open-set DA and UniDA approaches treat all novel categories as one unified unknown class and attempt to detect this unknown class during the training process. However, the features of the novel categories learned by these methods are not discriminative. This limits the applicability of UDA in the further classification of these novel categories into their original categories, rather than assigning them to a single unified class. In this paper, we propose a self-labeling framework to cluster all target samples, including those in the ''unknown'' categories. We train the network to learn the representations of target samples via self-supervised learning (SSL) and to identify the seen and unseen (novel) target-sample categories simultaneously by maximizing the mutual information between labels and input data. We evaluated our approach under different DA settings and concluded that our method generally outperformed existing ones by a wide margin.

GRAPH NEURAL NETWORK BASED OPEN-SET DOMAIN ADAPTATION

Abstract. Owing to the presence of many sensors and geographic/seasonal variations, domain adaptation is an important topic in remote sensing. However, most domain adaptation methods focus on close-set adaptation, i.e., they assume that the source and target domains share the same label space. This assumption often does not hold in practice, as there can be previously unseen classes in the target domain. To circumnavigate this issue, we propose a method for open set domain adaptation, where the target domain contains additional unknown classes that are not present in the source domain. To improve the model’s generalization ability, we propose a Progressive Weighted Graph Learning (PWGL) method. The proposed method exploits graph neural networks in aggregating similar samples across source and target domains. The progressive strategy gradually separates the unknown samples apart from known samples and upgrades the source domain by incorporating the pseudolabeled known target samples. The weighted adversarial learning promotes the alignment of known classes across different domains and rejects the unknown class. The experiments performed on a multi-city dataset show the effectiveness of the proposed approach.

Interactive dual adversarial neural network framework: An open-set domain adaptation intelligent fault diagnosis method of rotating machinery

The domain-adaptation technique has been proven to be able to resolve the fault diagnosis under various working conditions. Most research presumes that the health states in the source domain are consistent with the target domain. However, open-set domain adaptation problem that contains the unknown states in testing process remains unexplored. Here we propose an interactive dual adversarial neural network (IDANN) for this problem. First, a closed-set domain adversarial network is trained to obtain the weight of each target instance. Then, an open-set domain adversarial network is trained by importing the weighted unknown classification items and entropy minimization techniques. Through a series of interactive training, the IDANN can not only distinguish the unknown instances but also assign known instances to corresponding classes. Two experiment cases validate the effectiveness of the proposed IDANN method. The comparison results suggest that the proposed method can achieve superior performance in open-set domain adaptation problems.