- Research Article
2337
- 10.1016/j.neucom.2018.05.083
Deep visual domain adaptation: A survey
- May 30, 2018
- Neurocomputing
- Mei Wang + 1 more +1
Deep visual domain adaptation: A survey
This research paper delves into leveraging Machine Learning (ML) for precise localization in GPSchallenged environments like urban canyons, addressing the complexities of time-varying signal propagation types, where transient obstructions, such as vehicles, can modify the channel state information (CSI) over time. It presents a novel approach termed semantic localization, which recognizes signal propagation conditions as semantic elements, incorporating them into the localization framework to enhance both accuracy and resilience. To tackle the issue of diverse CSIs at each location and the extensive need for labeled data, the paper proposes a multi-task deep domain adaptation (DA) strategy. This approach trains neural networks using a limited set of labeled data complemented by a vast array of unlabeled samples, coupled with innovative scenario adaptive learning techniques for optimal representation learning and knowledge transfer. Employing Bayesian theory for the efficient management of task importance weights minimizes the necessity for laborious parameter tuning. By making certain assumptions, the study introduces a deep homoscedastic DA method for enhanced joint task efficacy. Through detailed simulations using a 3D ray tracing dataset, the paper evidences that the integration of environmental semantics and the advanced DA localization techniques markedly elevates the precision of localization in various demanding settings.
Deep visual domain adaptation: A survey
Deep visual domain adaptation: A survey
A Multi-Source Domain Adaptation Method for Bearing Fault Diagnosis with Dynamically Similarity Guidance on Incomplete Data
In actual industrial scenarios, collecting a complete dataset with all fault categories under the same conditions is challenging, leading to a loss in fault category knowledge in single-source domains. Deep learning domain adaptation methods face difficulties in multi-source scenarios due to insufficient labeled data and significant distribution differences, hindering domain-specific knowledge transfer and reducing fault diagnosis efficiency. To address these issues, the Dynamic Similarity-guided Multi-source Domain Adaptation Network (DS-MDAN) is proposed. This method leverages incomplete data from multiple-source domains to address distribution disparities in deep domain adaptation. It enhances diagnostic performance in the target domain by transferring knowledge across diverse domains. DS-MDAN uses convolution kernels of different scales to extract multi-scale feature information and achieves feature fusion through upsampling and operations like addition and concatenation. Adversarial training with domain and fault classifiers optimizes feature extraction for widely applicable representations. The similarity between source and target domain data is calculated based on features extracted by a shared-weight network, dynamically adjusting the contribution of different source domain data to minimize distribution differences. Finally, matched source and target domain samples are mapped to the same feature space for fault diagnosis. Experimental validation on various bearing fault datasets shows that DS-MDAN improves performance in multiple fault diagnosis tasks, increasing accuracy and demonstrating good generalization capabilities.
Read moreDeep domain adaptation with manifold aligned label transfer
We propose a novel deep learning domain adaptation method that performs transductive learning from the source domain to the target domain based on cluster matching between the source and target features. The proposed method combines Adaptive Batch Normalization and Locality Preserving Projection-based subspace alignment on deep features to produce a common feature space for label transfer. Adaptive Batch Normalization automatically conditions the features from the source/target domain by normalizing the activations in each layer of our network. Following Manifold Subspace Alignment, we cluster the data in each domain using Gaussian Mixture Model clustering in feature space. The clusters are matched between domains to transfer labels from the closest source cluster to each target cluster. The transfer labels are compared to the network prediction, and the samples with consistent labels are used to adapt the network on the target domain. The proposed manifold-guided label transfer method produces state-of-the-art results for deep adaptation on digit recognition datasets. Furthermore, we perform domain adaptation on remote sensing datasets.
Read moreTwo-Branch Attention Adversarial Domain Adaptation Network for Hyperspectral Image Classification
Recent studies have shown that deep domain adaptation (DA) techniques have good performance on cross-domain hyperspectral image (HSI) classification problems. However, most existing deep HSI DA approaches directly use deep networks to extract features from the data, which ignores the detailed information of HSI in spectral and spatial dimensions. To effectively exploit the spectral–spatial joint information for DA of HSIs, we propose a two-branch attention adversarial DA (TAADA) network in this article. In the TAADA network, a two-branch feature extraction (TBFE) subnetwork is first designed as a generator to extract the attention-based spectral–spatial features. Then, a discriminator based on two classifiers with the multilayer FC-BN-ReLU-Dropout structure is constructed. Based on adversarial learning between the generator and the discriminator, the ability of discriminative feature extraction and cross-domain classification is improved simultaneously. Finally, the TAADA network can adjust the distribution between the source and target domains and extract domain-invariant features. Experimental results on three cross-scene HSI classification tasks show that our proposed TAADA outperforms some existing DA methods.
Read moreAn attention guided joint distribution domain adaptation method for fault diagnosis
Deep domain adaptation (DA) methods are widely employed to address data distribution inconsistencies in engineering scenarios. Although DA methods have demonstrated effectiveness in cross-domain fault diagnosis, several limitations remain. Most studies primarily focus on aligning marginal distributions while often overlooking the alignment of conditional distributions, which can lead to inaccurate class alignment. To enhance the alignment between source and target domains and reduce class confusion at the decision boundary, an attention-guided joint distribution domain adaptation method (DRCDA) is proposed. First, domain adversarial neural networks (DANN) are utilized to align marginal distributions between the source and target domains. Next, a class alignment module is introduced to align conditional distributions by computing inter-class distances and applying multi-kernel maximum mean discrepancy (MK-MMD). This method incorporates pseudo-labels from the target domain alongside true labels from the source domain, facilitating compact clustering of similar features while ensuring separation between different classes. Additionally, an adaptive threshold pseudo-labeling strategy is designed to address the issue of low-quality pseudo-labels in the target domain. To further mitigate negative transfer effects, an inter-domain attention module is proposed to explore transferable contextual information and model the correlation between the source and target domains. The effectiveness of the DRCDA method is evaluated using two bearing datasets and one gearbox dataset. Experimental results confirm the superiority of the proposed method in cross-condition diagnostic tasks.
Read moreClass Consistency Driven Unsupervised Deep Adversarial Domain Adaptation
In unsupervised deep domain adaptation (DA), the use of adversarial domain classifiers is popular in learning a shared feature space which reduces the distributions gap for a pair of source (with training data) and target (with only test data) domains. In the new space, a classifier trained on source training data is expected to generalize well for the target domain samples. We hypothesize that such a feature space obtained by aligning the domains globally ignores the category level feature distributions. This, in turn, leads to erroneous mapping for fine-grained classes. Besides, the discriminativeness of the shared space is not explicitly addressed. In order to resolve both the issues, we propose a novel adversarial approach which judiciously refines the space learned by the domain classifier by incorporating class level information. We follow an ensemble classifiers based approach to model the source domain and introduce a novel consistency constrain on the classifier's outcomes when evaluated on a held-out set of target domain samples. We further leverage the ensemble learning strategy during the inference, as opposed to the existing single classifier based methods. We find that our deep DA model is capable of producing a compact and better domain aligned feature space. Experimental results obtained on the Office-Home, Office-CalTech, MNIST-USPS, and a remote sensing dataset confirm the superiority of the proposed approach.
Read moreDomain Conditioned Adaptation Network
Tremendous research efforts have been made to thrive deep domain adaptation (DA) by seeking domain-invariant features. Most existing deep DA models only focus on aligning feature representations of task-specific layers across domains while integrating a totally shared convolutional architecture for source and target. However, we argue that such strongly-shared convolutional layers might be harmful for domain-specific feature learning when source and target data distribution differs to a large extent. In this paper, we relax a shared-convnets assumption made by previous DA methods and propose a Domain Conditioned Adaptation Network (DCAN), which aims to excite distinct convolutional channels with a domain conditioned channel attention mechanism. As a result, the critical low-level domain-dependent knowledge could be explored appropriately. As far as we know, this is the first work to explore the domain-wise convolutional channel activation for deep DA networks. Moreover, to effectively align high-level feature distributions across two domains, we further deploy domain conditioned feature correction blocks after task-specific layers, which will explicitly correct the domain discrepancy. Extensive experiments on three cross-domain benchmarks demonstrate the proposed approach outperforms existing methods by a large margin, especially on very tough cross-domain learning tasks.
Read moreDeep domain adaptation with ordinal regression for pain assessment using weakly-labeled videos
Deep domain adaptation with ordinal regression for pain assessment using weakly-labeled videos
Cross-domain retinopathy classification with optical coherence tomography images via a novel deep domain adaptation method.
Deep learning based retinopathy classification with optical coherence tomography (OCT) images has recently attracted great attention. However, existing deep learning methods fail to work well when training and testing datasets are different due to the general issue of domain shift between datasets caused by different collection devices, subjects, imaging parameters, etc. To address this practical and challenging issue, we propose a novel deep domain adaptation (DDA) method to train a model on a labeled dataset and adapt it to an unlabelled dataset (collected under different conditions). It consists of two modules for domain alignment, that is, adversarial learning and entropy minimization. We conduct extensive experiments on three public datasets to evaluate the performance of the proposed method. The results indicate that there are large domain shifts between datasets, resulting a poor performance for conventional deep learning methods. The proposed DDA method can significantly outperform existing methods for retinopathy classification with OCT images. It achieves retinopathy classification accuracies of 0.915, 0.959 and 0.990 under three cross-domain (cross-dataset) scenarios. Moreover, it obtains a comparable performance with human experts on a dataset where no labeled data in this dataset have been used to train the proposed DDA method. We have also visualized the learnt features by using the t-distributed stochastic neighbor embedding (t-SNE) technique. The results demonstrate that the proposed method can learn discriminative features for retinopathy classification.
Read moreDeep learning-based cross-domain adaptation for gearbox fault diagnosis under variable speed conditions
Existing intelligent gearbox fault diagnosis approaches have two shortcomings: (a) their performance is mostly confined to manual handcrafted features, and (b) they follow a general assumption that the distribution of the data in the source domain (labeled data on which the model is trained) is similar to the target domain (unlabeled data on which the model is tested), which might not be the case in real-world applications. Substantial human expertise and domain knowledge is required for manual feature extraction, and moreover, deploying the same model for a target domain whose distribution is different from the source domain would lead to poor generalization. Since deep learning methods can automatically learn high dimensional feature representations from raw measurement data, this paper proposes a novel deep learning-based domain adaptation (DA) method for gearbox fault diagnosis under significant speed variations. A deep convolutional neural network is used as the main architecture. The paper proposes to minimize the summation of cross-entropy loss (between the labeled source domain data) and maximum mean discrepancy loss (between the labeled source and unlabeled target datasets) simultaneously to adapt the source domain model to be applied in the target domain. The proposed deep learning DA approach is evaluated using experimental data from a gearbox under variable speeds and multiple health conditions. An appropriate benchmarking with both traditional machine learning methods and other DA methods demonstrate the superiority of the proposed method.
Read moreDeep Domain Adaptation for Vulnerable Code Function Identification
Due to the ubiquity of computer software, software vulnerability detection (SVD) has become crucial in the software industry and in the field of computer security. Two significant issues in SVD arise when using machine learning, namely: i) how to learn automatic features that can help improve the predictive performance of vulnerability detection and ii) how to overcome the scarcity of labeled vulnerabilities in projects that require the laborious labeling of code by software security experts. In this paper, we address these two crucial concerns by proposing a novel architecture which leverages deep domain adaptation with automatic feature learning for software vulnerability identification. Based on this architecture, we keep the principles and reapply the state-of-the-art deep domain adaptation methods to indicate that deep domain adaptation for SVD is plausible and promising. Moreover, we further propose a novel method named Semi-supervised Code Domain Adaptation Network (SCDAN) that can efficiently utilize and exploit information carried in unlabeled target data by considering them as the unlabeled portion in a semi-supervised learning context. The proposed SCDAN method enforces the clustering assumption, which is a key principle in semi-supervised learning. The experimental results using six real-world software project datasets show that our SCDAN method and the baselines using our architecture have better predictive performance by a wide margin compared with the Deep Code Network (VulDeePecker) method without domain adaptation. Also, the proposed SCDAN significantly outperforms the DIRT-T which to the best of our knowledge is currently the-state-of-the-art method in deep domain adaptation and other baselines.
Read moreDeep Residual Correction Network for Partial Domain Adaptation.
Deep domain adaptation methods have achieved appealing performance by learning transferable representations from a well-labeled source domain to a different but related unlabeled target domain. Most existing works assume source and target data share the identical label space, which is often difficult to be satisfied in many real-world applications. With the emergence of big data, there is a more practical scenario called partial domain adaptation, where we are always accessible to a more large-scale source domain while working on a relative small-scale target domain. In this case, the conventional domain adaptation assumption should be relaxed, and the target label space tends to be a subset of the source label space. Intuitively, reinforcing the positive effects of the most relevant source subclasses and reducing the negative impacts of irrelevant source subclasses are of vital importance to address partial domain adaptation challenge. This paper proposes an efficiently-implemented Deep Residual Correction Network (DRCN) by plugging one residual block into the source network along with the task-specific feature layer, which effectively enhances the adaptation from source to target and explicitly weakens the influence from the irrelevant source classes. Specifically, the plugged residual block, which consists of several fully-connected layers, could deepen basic network and boost its feature representation capability correspondingly. Moreover, we design a weighted class-wise domain alignment loss to couple two domains by matching the feature distributions of shared classes between source and target. Comprehensive experiments on partial, traditional and fine-grained cross-domain visual recognition demonstrate that DRCN is superior to the competitive deep domain adaptation approaches.
Read moreDeep sample clustering domain adaptation for breast histopathology image classification
Deep sample clustering domain adaptation for breast histopathology image classification
A new deep domain adaptation method with joint adversarial training for online detection of bearing early fault
A new deep domain adaptation method with joint adversarial training for online detection of bearing early fault
Deep Discriminative Feature Learning for Domain Adaptation
Recent advancements have been seen in Deep domain adaptation field, which helps transfer knowledge from a source domain to a related but different target domain, greatly reducing the cost of manual annotation and successfully learning domain invariant features. However, most existing deep domain adaptation methods only align source and target domain distributions, neglecting the class structure information in the source domain, and ultimately leading to domain confusion. To address this issue, we propose a new model for deep domain adaptation, which can simultaneously achieve domain alignment and discriminative feature learning. Specifically, apart from performing domain-invariant embeddings with MMD metric, we utilize a center loss to construct class structure, so as to enhance inter-class separability and intra-class compactness. In addition, our model is effective and easy to implement, compared to other methods. Extensive experiments conducted on two benchmark datasets verify that our model has superior performance over state-of-the-art methods.
Read more