Rethinking Maximum Mean Discrepancy for Visual Domain Adaptation

Cross-Domain Invariant Feature Absorption and Domain-Specific Feature Retention for Domain Incremental Chest X-Ray Classification

Chest X-ray (CXR) images have been widely adopted in clinical care and pathological diagnosis in recent years. Some advanced methods on CXR classification task achieve impressive performance by training the model statically. However, in the real clinical environment, the model needs to learn continually and this can be viewed as a domain incremental learning (DIL) problem. Due to large domain gaps, DIL is faced with catastrophic forgetting. Therefore, in this paper, we propose a Cross-domain invariant feature absorption and Domain-specific feature retention (CaD) framework. To be specific, we adopt a Cross-domain Invariant Feature Absorption (CIFA) module to learn the domain invariant knowledge and a Domain-Specific Feature Retention (DSFR) module to learn the domain-specific knowledge. The CIFA module contains the C(lass)-adapter and an absorbing strategy is used to fuse the common features among different domains. The DSFR module contains the D(omain)-adapter for each domain and it connects to the network in parallel independently to prevent forgetting. A multi-label contrastive loss (MLCL) is used in the training process and improves the class distinctiveness within each domain. We leverage publicly available large-scale datasets to simulate domain incremental learning scenarios, extensive experimental results substantiate the effectiveness of our proposed methods and it has reached state-of-the-art performance.

Addressing cross-population domain shift in chest X-ray classification through supervised adversarial domain adaptation

Medical image analysis, empowered by artificial intelligence (AI), plays a crucial role in modern healthcare diagnostics. However, the effectiveness of machine learning models hinges on their ability to generalize to diverse patient populations, presenting domain shift challenges. This study explores the domain shift problem in chest X-ray classification, focusing on cross-population variations, especially in underrepresented groups. We analyze the impact of domain shifts across three population datasets acting as sources using a Nigerian chest X-ray dataset acting as the target. Model performance is evaluated to assess disparities between source and target populations, revealing large discrepancies when the models trained on a source were applied to the target domain. To address with the evident domain shift among the populations, we propose a supervised adversarial domain adaptation (ADA) technique. The feature extractor is first trained on the source domain using a supervised loss function in ADA. The feature extractor is then frozen, and an adversarial domain discriminator is introduced to distinguish between the source and target domains. Adversarial training fine-tunes the feature extractor, making features from both domains indistinguishable, thereby creating domain-invariant features. The technique was evaluated on the Nigerian dataset, showing significant improvements in chest X-ray classification performance. The proposed model achieved a 90.08% accuracy and a 96% AUC score, outperforming existing approaches such as multi-task learning (MTL) and continual learning (CL). This research highlights the importance of developing domain-aware models in AI-driven healthcare, offering a solution to cross-population domain shift challenges in medical imaging.

Transferability-Guided Cross-Domain Cross-Task Transfer Learning

We propose two novel transferability metrics fast optimal transport-based conditional entropy (F-OTCE) and joint correspondence OTCE (JC-OTCE) to evaluate how much the source model (task) can benefit the learning of the target task and to learn more generalizable representations for cross-domain cross-task transfer learning. Unlike the original OTCE metric that requires evaluating the empirical transferability on auxiliary tasks, our metrics are auxiliary-free such that they can be computed much more efficiently. Specifically, F-OTCE estimates transferability by first solving an optimal transport (OT) problem between source and target distributions and then uses the optimal coupling to compute the negative conditional entropy (NCE) between the source and target labels. It can also serve as an objective function to enhance downstream transfer learning tasks including model finetuning and domain generalization (DG). Meanwhile, JC-OTCE improves the transferability accuracy of F-OTCE by including label distances in the OT problem, though it incurs additional computation costs. Extensive experiments demonstrate that F-OTCE and JC-OTCE outperform state-of-the-art auxiliary-free metrics by 21.1% and 25.8%, respectively, in correlation coefficient with the ground-truth transfer accuracy. By eliminating the training cost of auxiliary tasks, the two metrics reduce the total computation time of the previous method from 43 min to 9.32 and 10.78 s, respectively, for a pair of tasks. When applied in the model finetuning and DG tasks, F-OTCE shows significant improvements in the transfer accuracy in few-shot classification experiments, with up to 4.41% and 2.34% accuracy gains, respectively.

Deep Multiview Module Adaption Transfer Network for Subject-Specific EEG Recognition

Transfer learning is one of the popular methods to solve the problem of insufficient data in subject-specific electroencephalogram (EEG) recognition tasks. However, most existing approaches ignore the difference between subjects and transfer the same feature representations from source domain to different target domains, resulting in poor transfer performance. To address this issue, we propose a novel subject-specific EEG recognition method named deep multiview module adaption transfer (DMV-MAT) network. First, we design a universal deep multiview (DMV) network to generate different types of discriminative features from multiple perspectives, which improves the generalization performance by extensive feature sets. Second, module adaption transfer (MAT) is designed to evaluate each module by the feature distributions of source and target samples, which can generate an optimal weight sharing strategy for each target subject and promote the model to learn domain-invariant and domain-specific features simultaneously. We conduct extensive experiments in two EEG recognition tasks, i.e., motor imagery (MI) and seizure prediction, on four datasets. Experimental results demonstrate that the proposed method achieves promising performance compared with the state-of-the-art methods, indicating a feasible solution for subject-specific EEG recognition tasks. Implementation codes are available at https://github.com/YangLibuaa/DMV-MAT.

Deep Label Propagation with Nuclear Norm Maximization for Visual Domain Adaptation

Domain adaptation aims to leverage abundant label information from a source domain to an unlabeled target domain with two different distributions. Existing methods usually rely on a classifier to generate high-quality pseudo-labels for the target domain, facilitating the learning of discriminative features. Label propagation (LP), as an effective classifier, propagates labels from the source domain to the target domain by designing a smooth function over a similarity graph, which represents structural relationships among data points in feature space. However, LP has not been thoroughly explored in deep neural network-based domain adaptation approaches. Additionally, the probability labels generated by LP are low-confident and LP is sensitive to class imbalance problem. To address these problems, we propose a novel approach for domain adaptation named deep label propagation with nuclear norm maximization (DLP-NNM). Specifically, we employ the constraint of nuclear norm maximization to enhance both label confidence and class diversity in LP and propose an efficient algorithm to solve the corresponding optimization problem. Subsequently, we utilize the proposed LP to guide the classifier layer in a deep discriminative adaptation network using the cross-entropy loss. As such, the network could produce more reliable predictions for the target domain, thereby facilitating more effective discriminative feature learning. Extensive experimental results on three cross-domain benchmark datasets demonstrate that the proposed DLP-NNM surpasses existing state-of-the-art domain adaptation approaches.

Domain Adaptation Based on Semi-Supervised Cross-Domain Mean Discriminative Analysis and Kernel Transfer Extreme Learning Machine

Good data feature representation and high precision classifiers are the key steps for pattern recognition. However, when the data distributions between testing samples and training samples do not match, the traditional feature extraction methods and classification models usually degrade. In this paper, we propose a domain adaptation approach to handle this problem. In our method, we first introduce cross-domain mean approximation (CDMA) into semi-supervised discriminative analysis (SDA) and design semi-supervised cross-domain mean discriminative analysis (SCDMDA) to extract shared features across domains. Secondly, a kernel extreme learning machine (KELM) is applied as a subsequent classifier for the classification task. Moreover, we design a cross-domain mean constraint term on the source domain into KELM and construct a kernel transfer extreme learning machine (KTELM) to further promote knowledge transfer. Finally, the experimental results from four real-world cross-domain visual datasets prove that the proposed method is more competitive than many other state-of-the-art methods.

Self-Supervised Autoregressive Domain Adaptation for Time Series Data

Unsupervised domain adaptation (UDA) has successfully addressed the domain shift problem for visual applications. Yet, these approaches may have limited performance for time series data due to the following reasons. First, they mainly rely on the large-scale dataset (i.e., ImageNet) for source pretraining, which is not applicable for time series data. Second, they ignore the temporal dimension on the feature space of the source and target domains during the domain alignment step. Finally, most of the prior UDA methods can only align the global features without considering the fine-grained class distribution of the target domain. To address these limitations, we propose a SeLf-supervised AutoRegressive Domain Adaptation (SLARDA) framework. In particular, we first design a self-supervised (SL) learning module that uses forecasting as an auxiliary task to improve the transferability of source features. Second, we propose a novel autoregressive domain adaptation technique that incorporates temporal dependence of both source and target features during domain alignment. Finally, we develop an ensemble teacher model to align class-wise distribution in the target domain via a confident pseudo labeling approach. Extensive experiments have been conducted on three real-world time series applications with 30 cross-domain scenarios. The results demonstrate that our proposed SLARDA method significantly outperforms the state-of-the-art approaches for time series domain adaptation. Our source code is available at: https://github.com/mohamedr002/SLARDA.

A New Progressive Multisource Domain Adaptation Network With Weighted Decision Fusion

Multisource unsupervised domain adaptation (MUDA) is an important and challenging topic for target classification with the assistance of labeled data in source domains. When we have several labeled source domains, it is difficult to map all source domains and target domain into a common feature space for classifying the targets well. In this article, a new progressive multisource domain adaptation network (PMSDAN) is proposed to further improve the classification performance. PMSDAN mainly consists of two steps for distribution alignment. First, the multiple source domains are integrated as one auxiliary domain to match the distribution with the target domain. By doing this, we can generally reduce the distribution discrepancy between each source and target domains, as well as the discrepancy between different source domains. It can efficiently explore useful knowledge from the integrated source domain. Second, to mine assistance knowledge from each source domain as much as possible, the distribution of the target domain is separately aligned with that of each source domain. A weighted fusion method is employed to combine the multiple classification results for making the final decision. In the optimization of domain adaption, weighted hybrid maximum mean discrepancy (WHMMD) is proposed, and it considers both the interclass and intraclass discrepancies. The effectiveness of the proposed PMSDAN is demonstrated in the experiments comparing with some state-of-the-art methods.