Cross-subject mental workload recognition using bi-classifier domain adversarial learning

To deploy Electroencephalogram (EEG) based Mental Workload Recognition (MWR) systems in the real world, it is crucial to develop general models that can be applied across subjects. Previous studies have utilized domain adaptation to mitigate inter-subject discrepancies in EEG data distributions. However, they have focused on reducing global domain discrepancy, while neglecting local workload-categorical domain divergence. This degrades the workload-discriminating ability of subject-invariant features. To deal with this problem, we propose a novel joint category-wise and domain-wise alignment Domain Adaptation (cdaDA) algorithm, using bi-classifier learning and domain discriminative adversarial learning. The bi-classifier learning approach is adopted to address the similarities and differences between categories, helping to align EEG data within the same mental workload categories. Additionally, the domain discriminative adversarial learning technique is adopted to consider global domain information by minimizing global domain discrepancy. By integrating both local category information and global domain information, the cdaDA model performs a coarse-to-fine alignment and achieves promising cross-subject MWR results.

A small-scale data driven and graph neural network based toxicity prediction method of compounds

Toxicity prediction is crucial in drug discovery, helping identify safe compounds and reduce development risks. However, the lack of known toxicity data for most compounds is a major challenge. Recently, data-driven models have gained attention as a more efficient alternative to traditional in vivo and in vitro experiments. In this paper, we propose a small-scale, data-driven toxicity prediction method based on Graph Neural Network (GNN). We introduce a joint learning strategy for multiple toxicity types and construct a graph-based model, JLGCN-MTT, to improve prediction accuracy. In addition, we integrate a transfer learning strategy that leverages data from multiple toxicity types, allowing the model to make reliable predictions even when data for a specific toxicity type is limited. We conducted experiments using data from 3566 compounds in the Tox21 dataset, which contains 12 types of toxicity-related bioactivity data. The experimental results show that JLGCN-MTT outperforms traditional machine learning methods and single-task GNN in all 12 toxicity prediction tasks, with AUC improving by over 10% in 11 tasks. For small-scale data with 50, 100, and 300 training samples, the AUC improved in all cases, with the highest improvement of 11% observed when the sample size was 50. These results demonstrate that the small-scale, data-driven toxicity prediction method we propose can achieve high prediction accuracy.

A novel deep transfer learning method based on explainable feature extraction and domain reconstruction

Although deep transfer learning has made significant progress, its "black-box" nature and unstable feature adaptation remain key obstacles. This study proposes a multi-stage deep transfer learning method, called XDTL, which combines explainable feature extraction and domain reconstruction to enhance the performance of target models. Specifically, the study first divides features into key and regular features through cross-validation and explainability analysis, then reconstructs the target domain using a seed replacement method based on key target samples, ultimately achieving deep transfer. Experimental results show that, compared to other methods, XDTL achieves an average improvement of 27.43 % in effectiveness, demonstrating superior performance and stronger explainability. This method offers new insights into addressing the explainability challenges in transfer learning and highlights its potential for broader applications across various tasks.

A deep transfer learning approach for Real-Time traffic conflict prediction with trajectory data

Recently, real-time traffic conflict prediction has drawn increasing attention due to its significant potential in proactive traffic safety systems. While various statistical and machine learning models have been developed for conflict prediction, transferability remains a fundamental issue across these models. Specifically, the predictive performance of a real-time conflict prediction model developed for a specific location can significantly decline when directly applied to a new location without any modifications, primarily due to substantial differences in traffic environments between these areas. To address this gap, this study proposed a novel deep transfer learning approach aimed at enhancing the transferability of real-time conflict prediction models. Initially, a real-time conflict prediction framework was designed utilizing trajectory data for merging areas with consideration of temporal variations in traffic flow characteristics. Subsequently, the Gated-Transformer, Fully Convolutional Networks (FCN), Long Short-Term Memory Fully Convolutional Networks (LSTM-FCN), and Multivariate Long Short-Term Memory Fully Convolutional Networks (MLSTM-FCN) were employed as backbone feature extraction networks to capture the hidden correlations between time-varying traffic flow characteristics and traffic conflicts. After that, an independent transfer learning architecture was established to assess the similarity of the distribution of traffic flow characteristics at different locations, based on the maximum mean discrepancy criteria. For empirical evaluation, merging areas from the exiD dataset were differentiated into source and target domains. The results demonstrated that the Gated-Transformer model outperforms other baseline models (FCN, LSTM-FCN and MLSTM-FCN) in both feature extraction and predictive performance, achieving an F1 score of 0.864 and an area under the curve (AUC) of 0.980. Furthermore, the transfer learning architecture can substantially enhance the predictive performance of a model trained in the source domain when applied to the target domain. In particular, the F1 score and AUC for the Gated-Transformer model improved by 11.9% and 10.2%, respectively, after incorporating the transfer learning architecture. Finally, the optimal values of key model parameters, including the sliding time window (6 s) and the prewarning time (5 s), were recommended for practical applications through sensitivity analysis. This study illustrates the potential of the deep transfer learning approach as a reliable and effective alternative to improve the transferability of real-time conflict prediction models. Additionally, results from this study can offer valuable insights for practical applications in traffic safety warning systems, particularly in vehicle-to-infrastructure traffic environments.

Dual-view global and local category-attentive domain alignment for unsupervised conditional adversarial domain adaptation

Conditional adversarial domain adaptation (CADA) is one of the most commonly used unsupervised domain adaptation (UDA) methods. CADA introduces multimodal information to the adversarial learning process to align the distributions of the labeled source domain and unlabeled target domain with mode match. However, CADA provides wrong multimodal information for challenging target features due to utilizing classifier predictions as the multimodal information, leading to distribution mismatch and less robust domain-invariant features. Compared to the recent state-of-the-art UDA methods, CADA also faces poor discriminability on the target domain. To tackle these challenges, we propose a novel unsupervised CADA framework named dual-view global and local category-attentive domain alignment (DV-GLCA). Specifically, to mitigate distribution mismatch and acquire more robust domain-invariant features, we integrate dual-view information into conditional adversarial domain adaptation and then utilize the substantial feature disparity between the two perspectives to better align the multimodal structures of the source and target distributions. Moreover, to learn more discriminative features of the target domain based on dual-view conditional adversarial domain adaptation (DV-CADA), we further propose global category-attentive domain alignment (GCA). We combine coding rate reduction and dual-view centroid alignment in GCA to amplify inter-category domain discrepancies while reducing intra-category domain differences globally. Additionally, to address challenging ambiguous samples during the training phase, we propose local category-attentive domain alignment (LCA). We introduce a new way of using contrastive domain discrepancy in LCA to move ambiguous samples closer to the correct category. Our method demonstrates leading performance on five UDA benchmarks, with extensive experiments showcasing its effectiveness.

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.

AdamGraph: Adaptive Attention-Modulated Graph Network for EEG Emotion Recognition

The underlying time-variant and subject-specific brain dynamics lead to inconsistent distributions in electroencephalogram (EEG) topology and representations within and between individuals. However, current works primarily align the distributions of EEG representations, overlooking the topology variability in capturing the dependencies between channels, which may limit the performance of EEG emotion recognition. To tackle this issue, this article proposes an adaptive attention-modulated graph network (AdamGraph) to enhance the subject adaptability of EEG emotion recognition against connection variability and representation variability. Specifically, an attention-modulated graph connection module is proposed to explicitly capture the individual important relationships among channels adaptively. Through modulating the attention matrix of individual functional connections using spatial connections based on prior knowledge, the attention-modulated weights can be learned to construct individual connections adaptively, thereby mitigating individual differences. Besides, a deep node-graph representation learning module is designed to extract long-range interaction characteristics among channels and alleviate the over-smoothing problem of representations. Furthermore, a graph domain co-regularized learning module is imposed to tackle the individual distribution discrepancies in connection and representations across different domains. Extensive experiments on three public EEG emotion datasets, i.e., SEED, DREAMER, and MPED, validate the superior performance of AdamGraph compared with state-of-the-art methods.

Recognition of MI-EEG signals using extended-LSR-based inductive transfer learning

Introduction

Motor imagery electroencephalographic (MI-EEG) signal recognition is used in various brain-computer interface (BCI) systems. In most existing BCI systems, this identification relies on classification algorithms. However, generally, a large amount of subject-specific labeled training data is required to reliably calibrate the classification algorithm for each new subject. To address this challenge, an effective strategy is to integrate transfer learning into the construction of intelligent models, allowing knowledge to be transferred from the source domain to enhance the performance of models trained in the target domain. Although transfer learning has been implemented in EEG signal recognition, many existing methods are designed specifically for certain intelligent models, limiting their application and generalization.

Methods

To broaden application and generalization, an extended-LSR-based inductive transfer learning method is proposed to facilitate transfer learning across various classical intelligent models, including neural networks, Takagi-SugenoKang (TSK) fuzzy systems, and kernel methods.

Results and discussion

The proposed method not only promotes the transfer of valuable knowledge from the source domain to improve learning performance in the target domain when target domain training data are insufficient but also enhances application and generalization by incorporating multiple classic base models. The experimental results demonstrate the effectiveness of the proposed method in MI-EEG signal recognition.

DP-MP: a novel cross-subject fatigue detection framework with DANN-based prototypical representation and mix-up pairwise learning

Objective. Electroencephalography (EEG) is widely recognized as an effective method for detecting fatigue. However, practical applications of EEG for fatigue detection in real-world scenarios are often challenging, particularly in cases involving subjects not included in the training datasets, owing to bio-individual differences and noisy labels. This study aims to develop an effective framework for cross-subject fatigue detection by addressing these challenges.Approach. In this study, we propose a novel framework, termed DP-MP, for cross-subject fatigue detection, which utilizes a domain-adversarial neural network-based prototypical representation in conjunction with Mix-up pairwise learning. Our proposed DP-MP framework aims to mitigate the impact of bio-individual differences by encoding fatigue-related semantic structures within EEG signals and exploring shared fatigue prototype features across individuals. Notably, to the best of our knowledge, this work is the first to conceptualize fatigue detection as a pairwise learning task, thereby effectively reducing the interference from noisy labels. Furthermore, we propose the Mix-up pairwise learning (MixPa) approach in the field of fatigue detection, which broadens the advantages of pairwise learning by introducing more diverse and informative relationships among samples.Main results. Cross-subject experiments were conducted on two benchmark databases, SEED-VIG and FTEF, achieving state-of-the-art performance with average accuracies of 88.14%and 97.41%, respectively. These promising results demonstrate our model's effectiveness and excellent generalization capability.Significance. This is the first time EEG-based fatigue detection has been conceptualized as a pairwise learning task, offering a novel perspective to this field. Moreover, our proposed DP-MP framework effectively tackles the challenges of bio-individual differences and noisy labels in the fatigue detection field and demonstrates superior performance. Our work provides valuable insights for future research, promoting the practical application of brain-computer interfaces for fatigue detection.

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.

An Efficient Graph Learning System for Emotion Recognition Inspired by the Cognitive Prior Graph of EEG Brain Network

Benefiting from the high-temporal resolution of electroencephalogram (EEG), EEG-based emotion recognition has become one of the hotspots of affective computing. For EEG-based emotion recognition systems, it is crucial to utilize state-of-the-art learning strategies to automatically learn emotion-related brain cognitive patterns from emotional EEG signals, and the learned stable cognitive patterns effectively ensure the robustness of the emotion recognition system. In this work, to realize the efficient decoding of emotional EEG, we propose a graph learning system [Graph Convolutional Network framework with Brain network initial inspiration and Fused attention mechanism (BF-GCN)] inspired by the brain cognitive mechanism to automatically learn graph patterns from emotional EEG and improve the performance of EEG emotion recognition. In the proposed BF-GCN, three graph branches, i.e., cognition-inspired functional graph branch, data-driven graph branch, and fused common graph branch, are first elaborately designed to automatically learn emotional cognitive graph patterns from emotional EEG signals. And then, the attention mechanism is adopted to further capture the brain activation graph patterns that are related to emotion cognition to achieve an efficient representation of emotional EEG signals. Essentially, the proposed BF-CGN model is a cognition-inspired graph learning neural network model, which utilizes the spectral graph filtering theory in the automatic learning and extracting of emotional EEG graph patterns. To evaluate the performance of the BF-GCN graph learning system, we conducted subject-dependent and subject-independent experiments on two public datasets, i.e., SEED and SEED-IV. The proposed BF-GCN graph learning system has achieved 97.44% (SEED) and 89.55% (SEED-IV) in subject-dependent experiments, and the results in subject-independent experiments have achieved 92.72% (SEED) and 82.03% (SEED-IV), respectively. The state-of-the-art performance indicates that the proposed BF-GCN graph learning system has a robust performance in EEG-based emotion recognition, which provides a promising direction for affective computing.

TODO-Net: Temporally Observed Domain Contrastive Network for 3-D Early Action Prediction

Early action prediction aiming to recognize which classes the actions belong to before they are fully conveyed is a very challenging task, owing to the insufficient discrimination information caused by the domain gaps among different temporally observed domains. Most of the existing approaches focus on using fully observed temporal domains to "guide" the partially observed domains while ignoring the discrepancies between the harder low-observed temporal domains and the easier highly observed temporal domains. The recognition models tend to learn the easier samples from the highly observed temporal domains and may lead to significant performance drops on low-observed temporal domains. Therefore, in this article, we propose a novel temporally observed domain contrastive network, namely, TODO-Net, to explicitly mine the discrimination information from the hard actions samples from the low-observed temporal domains by mitigating the domain gaps among various temporally observed domains for 3-D early action prediction. More specifically, the proposed TODO-Net is able to mine the relationship between the low-observed sequences and all the highly observed sequences belonging to the same action category to boost the recognition performance of the hard samples with fewer observed frames. We also introduce a temporal domain conditioned supervised contrastive (TD-conditioned SupCon) learning scheme to empower our TODO-Net with the ability to minimize the gaps between the temporal domains within the same action categories, meanwhile pushing apart the temporal domains belonging to different action classes. We conduct extensive experiments on two public 3-D skeleton-based activity datasets, and the results show the efficacy of the proposed TODO-Net.

Semantic Correlation Transfer for Heterogeneous Domain Adaptation

Heterogeneous domain adaptation (HDA) is expected to achieve effective knowledge transfer from a label-rich source domain to a heterogeneous target domain with scarce labeled data. Most prior HDA methods strive to align the cross-domain feature distributions by learning domain invariant representations without considering the intrinsic semantic correlations among categories, which inevitably results in the suboptimal adaptation performance across domains. Therefore, to address this issue, we propose a novel semantic correlation transfer (SCT) method for HDA, which not only matches the marginal and conditional distributions between domains to mitigate the large domain discrepancy, but also transfers the category correlation knowledge underlying the source domain to target by maximizing the pairwise class similarity across source and target. Technically, the domainwise and classwise centroids (prototypes) are first computed and aligned according to the feature embeddings. Then, based on the derived classwise prototypes, we leverage the cosine similarity of each two classes in both domains to transfer the supervised source semantic correlation knowledge among different categories to target effectively. As a result, the feature transferability and category discriminability can be simultaneously improved during the adaptation process. Comprehensive experiments and ablation studies on standard HDA tasks, such as text-to-image, image-to-image, and text-to-text, have demonstrated the superiority of our proposed SCT against several state-of-the-art HDA methods.

Multi-layer transfer learning algorithm based on improved common spatial pattern for brain-computer interfaces

In the application of brain-computer interface, the differences in imaging methods and brain structure between subjects hinder the effectiveness of decoding algorithms when applied on different subjects. Transfer learning has been designed to solve this problem. There have been many applications of transfer learning in motor imagery (MI), however the effectiveness is still limited due to the inconsistent domain alignment, lack of prominent data features and allocation of weights in trails. In this paper, a Multi-layer transfer learning algorithm based on improved Common Spatial Patterns (MTICSP) was proposed to solve these problems. Firstly, the source domain data and target domain data were aligned by Target Alignment (TA)method to reduce distribution differences between subjects. Secondly, the mean covariance matrix of the two classes was re-weighted by calculating the distance between the covariance matrix of each trial in the source domain and the target domain. Thirdly, the improved Common Spatial Patterns (CSP) by introducing regularization coefficient was proposed to further reduce the difference between source domain and target domain to extract features. Finally, the feature blocks of the source domain and target domain were aligned again by Joint Distribution Adaptation (JDA) method. Experiments on two public datasets in two transfer paradigms multi-source to single-target (MTS) and single-source to single-target (STS) verified the effectiveness of our proposed method. The MTS and STS in the 5-person dataset were 80.21% and 77.58%, respectively, and 80.10% and 73.91%, respectively, in the 9-person dataset. Experimental results also showed that the proposed algorithm was superior to other state-of-the-art algorithms. In addition, the generalization ability of our algorithm MTICSP was validated on the fatigue EEG dataset collected by ourselves, and obtained 94.83% and 87.41% accuracy in MTS and STS experiments respectively. The proposed method combines improved CSP with transfer learning to extract the features of source and target domains effectively, providing a new method for combining transfer learning with motor imagination.

Smooth-Guided Implicit Data Augmentation for Domain Generalization

The training process of a domain generalization (DG) model involves utilizing one or more interrelated source domains to attain optimal performance on an unseen target domain. Existing DG methods often use auxiliary networks or require high computational costs to improve the model's generalization ability by incorporating a diverse set of source domains. In contrast, this work proposes a method called Smooth-Guided Implicit Data Augmentation (SGIDA) that operates in the feature space to capture the diversity of source domains. To amplify the model's generalization capacity, a distance metric learning (DML) loss function is incorporated. Additionally, rather than depending on deep features, the suggested approach employs logits produced from cross entropy (CE) losses with infinite augmentations. A theoretical analysis shows that logits are effective in estimating distances defined on original features, and the proposed approach is thoroughly analyzed to provide a better understanding of why logits are beneficial for DG. Moreover, to increase the diversity of the source domain, a sampling-based method called smooth is introduced to obtain semantic directions from interclass relations. The effectiveness of the proposed approach is demonstrated through extensive experiments on widely used DG, object detection, and remote sensing datasets, where it achieves significant improvements over existing state-of-the-art methods across various backbone networks.

Heterogeneous Domain Adaptation With Generalized Similarity and Dissimilarity Regularization

Heterogeneous domain adaptation (HDA) aims to address the transfer learning problems where the source domain and target domain are represented by heterogeneous features. The existing HDA methods based on matrix factorization have been proven to learn transferable features effectively. However, these methods only preserve the original neighbor structure of samples in each domain and do not use the label information to explore the similarity and separability between samples. This would not eliminate the cross-domain bias of samples and may mix cross-domain samples of different classes in the common subspace, misleading the discriminative feature learning of target samples. To tackle the aforementioned problems, we propose a novel matrix factorization-based HDA method called HDA with generalized similarity and dissimilarity regularization (HGSDR). Specifically, we propose a similarity regularizer by establishing the cross-domain Laplacian graph with label information to explore the similarity between cross-domain samples from the identical class. And we propose a dissimilarity regularizer based on the inner product strategy to expand the separability of cross-domain labeled samples from different classes. For unlabeled target samples, we keep their neighbor relationship to preserve the similarity and separability between them in the original space. Hence, the generalized similarity and dissimilarity regularization is built by integrating the above regularizers to facilitate cross-domain samples to form discriminative class distributions. HGSDR can more efficiently match the distributions of the two domains both from the global and sample viewpoints, thereby learning discriminative features for target samples. Extensive experiments on the benchmark datasets demonstrate the superiority of the proposed method against several state-of-the-art methods.

Spatial transcriptomics-aided localization for single-cell transcriptomics with STALocator

Single-cell RNA-sequencing (scRNA-seq) techniques can measure gene expression at single-cell resolution but lack spatial information. Spatial transcriptomics (ST) techniques simultaneously provide gene expression data and spatial information. However, the data quality of the spatial resolution or gene coverage is still much lower than the quality of the single-cell transcriptomics data. To this end, we develop a ST-Aided Locator for single-cell transcriptomics (STALocator) to localize single cells to corresponding ST data. Applications on simulated data showed that STALocator performed better than other localization methods. When applied to the human brain and squamous cell carcinoma data, STALocator could robustly reconstruct the relative spatial organization of critical cell populations. Moreover, STALocator could enhance gene expression patterns for Slide-seqV2 data and predict genome-wide gene expression data for fluorescence in situ hybridization (FISH) and Xenium data, leading to the identification of more spatially variable genes and more biologically relevant Gene Ontology (GO) terms compared with the raw data. A record of this paper's transparent peer review process is included in the supplemental information.

A Cross-Machine Intelligent Fault Diagnosis Method with Small and Imbalanced Data Based on the ResFCN Deep Transfer Learning Model

Intelligent fault diagnosis (IFD) for mechanical equipment based on small and imbalanced datasets has been widely studied in recent years, with transfer learning emerging as one of the most promising approaches. Existing transfer learning-based IFD methods typically use data from different operating conditions of the same equipment as the source and target domains for the transfer learning process. However, in practice, it is often challenging to find identical equipment to obtain source domain data when diagnosing faults in the target equipment. These strict assumptions pose significant limitations on the application of IFD techniques in real-world industrial settings. Furthermore, the temporal characteristics of time-series monitoring data are often inadequately considered in existing methods. In this paper, we propose a cross-machine IFD method based on a residual full convolutional neural network (ResFCN) transfer learning model, which leverages the time-series features of monitoring data. By incorporating sliding window (SW)-based data segmentation, network pretraining, and model fine-tuning, the proposed method effectively exploits fault-associated general features in the source domain and learns domain-specific patterns that better align with the target domain, ultimately achieving accurate fault diagnosis for the target equipment. We design and implement three sets of experiments using two widely used public datasets. The results demonstrate that the proposed method outperforms existing approaches in terms of fault diagnosis accuracy and robustness.

LR Aerial Photo Categorization by Cross-Resolution Perceptual Knowledge Propagation

There are hundreds of high- and low-altitude earth observation satellites that asynchronously capture massive-scale aerial photographs every day. Generally, high-altitude satellites take low-resolution (LR) aerial pictures, each covering a considerably large area. In contrast, low-altitude satellites capture high-resolution (HR) aerial photos, each depicting a relatively small area. Accurately discovering the semantics of LR aerial photos is an indispensable technique in computer vision. Nevertheless, it is also a challenging task due to: 1) the difficulty to characterize human hierarchical visual perception and 2) the intolerable human resources to label sufficient training data. To handle these problems, a novel cross-resolution perceptual knowledge propagation (CPKP) framework is proposed, focusing on adapting the visual perceptual experiences deeply learned from HR aerial photos to categorize LR ones. Specifically, by mimicking the human vision system, a novel low-rank model is designed to decompose each LR aerial photo into multiple visually/semantically salient foreground regions coupled with the background nonsalient regions. This model can: 1) produce a gaze-shifting path (GSP) simulating human gaze behavior and 2) engineer the deep feature for each GSP. Afterward, a kernel-induced feature selection (FS) algorithm is formulated to obtain a succinct set of deep GSP features discriminative across LR and HR aerial photos. Based on the selected features, the labels from LR and HR aerial photos are collaboratively utilized to train a linear classifier for categorizing LR ones. It is worth emphasizing that, such a CPKP mechanism can effectively optimize the linear classifier training, as labels of HR aerial photos are acquired more conveniently in practice. Comprehensive visualization results and comparative study have validated the superiority of our approach.

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.

Lightweight Source-Free Domain Adaptation Based on Adaptive Euclidean Alignment for Brain-Computer Interfaces

For privacy protection of subjects in electroencephalogram (EEG)-based brain-computer interfaces (BCIs), using source-free domain adaptation (SFDA) for cross-subject recognition has proven to be highly effective. However, updating and storing a model trained on source subjects for each new subject can be inconvenient. This paper extends Euclidean alignment (EA) to propose adaptive Euclidean alignment (AEA), which learns a projection matrix to align the distribution of the target subject with the source subjects, thus eliminating domain drift issues and improving model classification performance of subject-independent BCIs. Combining the proposed AEA with various existing SFDA methods, such as SHOT, GSFDA, and NRC, this paper presents three new methods: AEA-SHOT, AEA-GSFDA, and AEA-NRC. In our experimental studies, these AEA-based SFDA methods were applied to four well-known deep learning models (i.e., EEGNet, Shallow ConvNet, Deep ConvNet, and MSFBCNN) on two motor imagery (MI) datasets, one event-related potential (ERP) dataset and one steady-state visual evoked potentials (SSVEP) dataset. The advanced cross-subject EEG classification performance demonstrates the efficacy of our proposed methods. For example, AEA-SHOT achieved the best average accuracy of 81.4% on the PhysioNet dataset.

Model-agnostic meta-learning for EEG-based inter-subject emotion recognition

Objective. Developing an efficient and generalizable method for inter-subject emotion recognition from neural signals is an emerging and challenging problem in affective computing. In particular, human subjects usually have heterogeneous neural signal characteristics and variable emotional activities that challenge the existing recognition algorithms from achieving high inter-subject emotion recognition accuracy.Approach. In this work, we propose a model-agnostic meta-learning algorithm to learn an adaptable and generalizable electroencephalogram-based emotion decoder at the subject's population level. Different from many prior end-to-end emotion recognition algorithms, our learning algorithms include a pre-training step and an adaptation step. Specifically, our meta-decoder first learns on diverse known subjects and then further adapts it to unknown subjects with one-shot adaptation. More importantly, our algorithm is compatible with a variety of mainstream machine learning decoders for emotion recognition.Main results. We evaluate the adapted decoders obtained by our proposed algorithm on three Emotion-EEG datasets: SEED, DEAP, and DREAMER. Our comprehensive experimental results show that the adapted meta-emotion decoder achieves state-of-the-art inter-subject emotion recognition accuracy and outperforms the classical supervised learning baseline across different decoder architectures.Significance. Our results hold promise to incorporate the proposed meta-learning emotion recognition algorithm to effectively improve the inter-subject generalizability in designing future affective brain-computer interfaces.

Simultaneously predicting SPAD and water content in rice leaves using hyperspectral imaging with deep multi-task regression and transfer component analysis

BACKGROUND

Water content and chlorophyll content are important indicators for monitoring rice growth status. Simultaneous detection of water content and chlorophyll content is of significance. Different varieties of rice show differences in phenotype, resulting in the difficulties of establishing a universal model. In this study, hyperspectral imaging was used to detect the Soil and Plant Analyzer Development (SPAD) values and water content of fresh rice leaves of three rice varieties (Jiahua 1, Xiushui 121 and Xiushui 134).

RESULTS

Both partial least squares regression and convolutional neural networks were used to establish single-task and multi-task models. Transfer component analysis (TCA) was used as transfer learning to learn the common features to achieve an approximate identical distribution between any two varieties. Single-task and multi-task models were also built using the features of the source domain, and these models were applied to the target domain. These results indicated that for models of each rice variety the prediction accuracy of most multi-task models was close to that of single-task models. As for TCA, the results showed that the single-task model achieved good performance for all transfer learning tasks.

CONCLUSION

Compared with the original model, good and differentiated results were obtained for the models using features learned by TCA for both the source domain and target domain. The multi-task models could be constructed to predict SPAD values and water content simultaneously and then transferred to another rice variety, which could improve the efficiency of model construction and realize rapid detection of rice growth indicators. © 2024 Society of Chemical Industry.

Wasserstein task embedding for measuring task similarities

Measuring similarities between different tasks is critical in a broad spectrum of machine learning problems, including transfer, multi-task, continual, and meta-learning. Most current approaches to measuring task similarities are architecture-dependent: (1) relying on pre-trained models, or (2) training networks on tasks and using forward transfer as a proxy for task similarity. In this paper, we leverage the optimal transport theory and define a novel task embedding for supervised classification that is model-agnostic, training-free, and capable of handling (partially) disjoint label sets. In short, given a dataset with ground-truth labels, we perform a label embedding through multi-dimensional scaling and concatenate dataset samples with their corresponding label embeddings. Then, we define the distance between two datasets as the 2-Wasserstein distance between their updated samples. Lastly, we leverage the 2-Wasserstein embedding framework to embed tasks into a vector space in which the Euclidean distance between the embedded points approximates the proposed 2-Wasserstein distance between tasks. We show that the proposed embedding leads to a significantly faster comparison of tasks compared to related approaches like the Optimal Transport Dataset Distance (OTDD). Furthermore, we demonstrate the effectiveness of our embedding through various numerical experiments and show statistically significant correlations between our proposed distance and the forward and backward transfer among tasks on a wide variety of image recognition datasets.

High-Density EMG Grip Force Estimation During Muscle Fatigue via Domain Adaptation

Myoelectric interfaces hold promise for enabling intuitive and natural control of prostheses and exoskeletons. Muscle fatigue, whether due to prolonged use or heavy weight loads, can alter the distribution of electromyographic (EMG) signals, leading to inconsistencies compared to non-fatigued conditions. This presents significant challenges for accurately decoding user intentions. We thus propose a novel estimation method based on domain adaptation to improve grip force estimation accuracy during muscle fatigue. Specifically, the proposed method integrates an adversarial training strategy and an end-to-end deep learning model to align EMG feature distributions across non-fatigue and fatigue states. A baseline model, whose structure was identical to the force estimation network of the proposed method, was used for comparison. Eight subjects performed non-fatigue and fatigue gripping tasks, and grip force estimations were compared with dynamometer gold standard measurements. Results demonstrate that root mean square error (RMSE) of the proposed model was 51.9% lower than that of the baseline model during muscle fatigue. The proposed method leverages labeled data in the non-fatigue domain and employs adversarial objectives to ensure the learned features are applicable to both domains, which eliminates the need to pause to collect force label data in the fatigue domain, expediting and simplifying the calibration process. This study has the potential to improve the ability of myoelectric interfaces during muscle fatigue to enable users to intuitively retrieve and grip objects over extended periods, ultimately improving independence and quality of life.

UMS2-ODNet: Unified-scale domain adaptation mechanism driven object detection network with multi-scale attention

Unsupervised domain adaptation techniques improve the generalization capability and performance of detectors, especially when the source and target domains have different distributions. Compared with two-stage detectors, one-stage detectors (especially YOLO series) provide better real-time capabilities and become primary choices in industrial fields. In this paper, to improve cross-domain object detection performance, we propose a Unified-Scale Domain Adaptation Mechanism Driven Object Detection Network with Multi-Scale Attention (UMS2-ODNet). UMS2-ODNet chooses YOLOv6 as the basic framework in terms of its balance between efficiency and accuracy. UMS2-ODNet considers the adaptation consistency across different scale feature maps, which tends to be ignored by existing methods. A unified-scale domain adaptation mechanism is designed to fully utilize and unify the discriminative information from different scales. A multi-scale attention module is constructed to further improve the multi-scale representation ability of features. A novel loss function is created to maintain the consistency of multi-scale information by considering the homology of the descriptions from the same latent feature. Multiply experiments are conducted on four widely used datasets. Our proposed method outperforms other state-of-the-art techniques, illustrating the feasibility and effectiveness of the proposed UMS2-ODNet.

Progressive Decision Boundary Shifting for Unsupervised Domain Adaptation

Unsupervised domain adaptation (UDA) is attracting more attention from researchers for boosting the task-specific generalization on target domain. It focuses on addressing the domain shift between the labeled source domain and the unlabeled target domain. Recent biclassifier-based UDA models perform category-level alignment to reduce domain shift, and meanwhile, self-training is used for improving the discriminability of target instances. However, the error accumulation problem of instances with high semantic uncertainty may cause discriminability degradation and category-level misalignment. To solve this issue, we design the progressive decision boundary shifting algorithm, where stable category information of target instances is explored for learning a discriminability structure on target domain. Specifically, we first model the semantic uncertainty of instances by progressively shifting decision boundaries of category. Then, we introduce the uncertainty decoupling in a contrastive manner, where the discriminative information is learned from the source domain for instance with low semantic uncertainty. Furthermore, we minimize the predictive entropy of instances with high semantic uncertainty to reduce their prediction confidence. Extensive experiments on three popular datasets show that our model outperforms the current state-of-the-art (SOTA) UDA methods.

An efficient method for identifying surface damage in hydraulic concrete buildings

Traditional hydraulic structures rely on manual visual inspection for apparent integrity, which is not only time-consuming and labour-intensive but also inefficient. The efficacy of deep learning models is frequently constrained by the size of available data, resulting in limited scalability and flexibility. Furthermore, the paucity of data diversity leads to a singular function of the model that cannot provide comprehensive decision support for improving maintenance measures. This paper proposes an efficacious methodology for identifying diverse apparent damages in hydraulic structures to address the limitations of existing technologies. The advanced features of apparent damage in hydraulic structures were elucidated by fine-tuning the top-level parameters of the lightweight pre-trained model, thereby mitigating the data dependency issue inherent in the model. Ensemble learning algorithms are employed to classify high-dimensional samples to enhance the accuracy and stability of the classification. However, ensemble learning algorithms are subject to time consuming issues when applied to high-dimensional datasets. To this end, we propose a robust discriminative feature selection model to identify the most salient features, thereby enhancing the performance of apparent damage recognition in hydraulic structures while concurrently reducing the inference time. The results demonstrated that the accuracies of this method in identifying crack, fracture, hole and normal structures were 87.65%, 87.82%, 96.99%, and 95.25%, respectively. This methodology exhibits significant applicability and practical value for the intelligent inspection of hydraulic structures.

A Lightweight Network with Domain Adaptation for Motor Imagery Recognition

Brain-computer interfaces (BCI) are an effective tool for recognizing motor imagery and have been widely applied in the motor control and assistive operation domains. However, traditional intention-recognition methods face several challenges, such as prolonged training times and limited cross-subject adaptability, which restrict their practical application. This paper proposes an innovative method that combines a lightweight convolutional neural network (CNN) with domain adaptation. A lightweight feature extraction module is designed to extract key features from both the source and target domains, effectively reducing the model's parameters and improving the real-time performance and computational efficiency. To address differences in sample distributions, a domain adaptation strategy is introduced to optimize the feature alignment. Furthermore, domain adversarial training is employed to promote the learning of domain-invariant features, significantly enhancing the model's cross-subject generalization ability. The proposed method was evaluated on an fNIRS motor imagery dataset, achieving an average accuracy of 87.76% in a three-class classification task. Additionally, lightweight experiments were conducted from two perspectives: model structure optimization and data feature selection. The results demonstrated the potential advantages of this method for practical applications in motor imagery recognition systems.

CrossIsoFun: predicting isoform functions using the integration of multi-omics data

MOTIVATION

Isoforms spliced from the same gene may carry distinct biological functions. Therefore, annotating functions at the isoform level provides valuable insights into the functional diversity of genomes. Since experimental approaches for determining isoform functions are time- and cost-demanding, computational methods have been proposed. In this case, multi-omics data integration helps enhance the model performance, providing complementary insights for isoform functions. However, current methods underperform in leveraging diverse omics data, primarily due to the limited power to integrate the heterogeneous feature domains. Besides, among the multi-omics data, isoform-isoform interactions (IIIs) are a key data source, as isoforms interact with each other to perform functions. Unfortunately, IIIs remain largely underutilized in isoform function predictions until now.

RESULTS

We introduce CrossIsoFun, a multi-omics data analysis framework for isoform function prediction. CrossIsoFun combines omics-specific and cross-omics learning for data integration and function prediction. In detail, CrossIsoFun uses a graph convolutional network (GCN) as the omics-specific classifier for each data source. The initial label predictions from GCNs are forwarded to the View Correlation Discovery Network (VCDN) and processed as a cross-omics integrative representation. The representation is then used to produce final predictions of isoform functions. In addition, an antoencoder within a cycle-consistency generative adversarial network (cycleGAN) is designed to generate IIIs from PPIs and thereby enrich the interactomics data. Our method outperforms the state-of-the-art methods on three tissue-naive datasets and 15 tissue-specific datasets with mRNA expression, sequence, and PPI data. The prediction of CrossIsoFun is further validated by its consistency with subcellular localization and isoform-level annotations with literature support.

AVAILABILITY AND IMPLEMENTATION

CrossIsoFun is freely available at https://github.com/genemine/CrossIsoFun.

Motion Cognitive Decoding of Cross-Subject Motor Imagery Guided on Different Visual Stimulus Materials

BACKGROUND

Motor imagery (MI) plays an important role in brain-computer interfaces, especially in evoking event-related desynchronization and synchronization (ERD/S) rhythms in electroencephalogram (EEG) signals. However, the procedure for performing a MI task for a single subject is subjective, making it difficult to determine the actual situation of an individual's MI task and resulting in significant individual EEG response variations during motion cognitive decoding.

METHODS

To explore this issue, we designed three visual stimuli (arrow, human, and robot), each of which was used to present three MI tasks (left arm, right arm, and feet), and evaluated differences in brain response in terms of ERD/S rhythms. To compare subject-specific variations of different visual stimuli, a novel cross-subject MI-EEG classification method was proposed for the three visual stimuli. The proposed method employed a covariance matrix centroid alignment for preprocessing of EEG samples, followed by a model agnostic meta-learning method for cross-subject MI-EEG classification.

RESULTS AND CONCLUSION

The experimental results showed that robot stimulus materials were better than arrow or human stimulus materials, with an optimal cross-subject motion cognitive decoding accuracy of 79.04%. Moreover, the proposed method produced robust classification of cross-subject MI-EEG signal decoding, showing superior results to conventional methods on collected EEG signals.

A temporal-spectral graph convolutional neural network model for EEG emotion recognition within and across subjects

EEG-based emotion recognition uses high-level information from neural activities to predict emotional responses in subjects. However, this information is sparsely distributed in frequency, time, and spatial domains and varied across subjects. To address these challenges in emotion recognition, we propose a novel neural network model named Temporal-Spectral Graph Convolutional Network (TSGCN). To capture high-level information distributed in time, spatial, and frequency domains, TSGCN considers both neural oscillation changes in different time windows and topological structures between different brain regions. Specifically, a Minimum Category Confusion (MCC) loss is used in TSGCN to reduce the inconsistencies between subjective ratings and predefined labels. In addition, to improve the generalization of TSGCN on cross-subject variation, we propose Deep and Shallow feature Dynamic Adversarial Learning (DSDAL) to calculate the distance between the source domain and the target domain. Extensive experiments were conducted on public datasets to demonstrate that TSGCN outperforms state-of-the-art methods in EEG-based emotion recognition. Ablation studies show that the mixed neural networks and our proposed methods in TSGCN significantly contribute to its high performance and robustness. Detailed investigations further provide the effectiveness of TSGCN in addressing the challenges in emotion recognition.

From Small Data Modeling to Large Language Model Screening: A Dual-Strategy Framework for Materials Intelligent Design

Small data in materials present significant challenges to constructing highly accurate machine learning models, severely hindering the widespread implementation of data-driven materials intelligent design. In this study, the Dual-Strategy Materials Intelligent Design Framework (DSMID) is introduced, which integrates two innovative methods. The Adversarial domain Adaptive Embedding Generative network (AAEG) transfers data between related property datasets, even with only 90 data points, enhancing material composition characterization and improving property prediction. Additionally, to address the challenge of screening and evaluating numerous alloy designs, the Automated Material Screening and Evaluation Pipeline (AMSEP) is implemented. This pipeline utilizes large language models with extensive domain knowledge to efficiently identify promising experimental candidates through self-retrieval and self-summarization. Experimental findings demonstrate that this approach effectively identifies and prepares new eutectic High Entropy Alloy (EHEA), notably Al14(CoCrFe)19Ni28, achieving an ultimate tensile strength of 1085 MPa and 24% elongation without heat treatment or extra processing. This demonstrates significantly greater plasticity and equivalent strength compared to the typical as-cast eutectic HEA AlCoCrFeNi2.1. The DSMID framework, combining AAEG and AMSEP, addresses the challenges of small data modeling and extensive candidate screening, contributing to cost reduction and enhanced efficiency of material design. This framework offers a promising avenue for intelligent material design, particularly in scenarios constrained by limited data availability.

Multi-source Selective Graph Domain Adaptation Network for cross-subject EEG emotion recognition

Affective brain-computer interface is an important part of realizing emotional human-computer interaction. However, existing objective individual differences among subjects significantly hinder the application of electroencephalography (EEG) emotion recognition. Existing methods still lack the complete extraction of subject-invariant representations for EEG and the ability to fuse valuable information from multiple subjects to facilitate the emotion recognition of the target subject. To address the above challenges, we propose a Multi-source Selective Graph Domain Adaptation Network (MSGDAN), which can better utilize data from different source subjects and perform more robust emotion recognition on the target subject. The proposed network extracts and selects the individual information specific to each subject, where public information refers to subject-invariant components from multi-source subjects. Moreover, the graph domain adaptation network captures both functional connectivity and regional states of the brain via a dynamic graph network and then integrates graph domain adaptation to ensure the invariance of both functional connectivity and regional states. To evaluate our method, we conduct cross-subject emotion recognition experiments on the SEED, SEED-IV, and DEAP datasets. The results demonstrate that the MSGDAN has superior classification performance.

An Adaptation-Aware Interactive Learning Approach for Multiple Operational Condition-Based Degradation Modeling

Although degradation modeling has been widely applied to use multiple sensor signals to monitor the degradation process and predict the remaining useful lifetime (RUL) of operating machinery units, three challenging issues remain. One challenge is that units in engineering cases usually work under multiple operational conditions, causing the distribution of sensor signals to vary over conditions. It remains unexplored to characterize time-varying conditions as a distribution shift problem. The second challenge is that sensor signal fusion and degradation status modeling are separated into two independent steps in most of the existing methods, which ignores the intrinsic correlation between the two parts. The last challenge is how to find an accurate health index (HI) of units using previous knowledge of degradation. To tackle these issues, this article proposes an adaptation-aware interactive learning (AAIL) approach for degradation modeling. First, a condition-invariant HI is developed to handle time-varying operation conditions. Second, an interactive framework based on the fusion and degradation model is constructed, which naturally integrates a supervised learner and an unsupervised learner. To estimate the model parameters of AAIL, we propose an interactive training algorithm that shares learned degradation and fusion information during the model training process. A case study that uses the degradation data set of aircraft engines demonstrates that the proposed AAIL outperforms related benchmark methods.

Enhancing Autism Spectrum Disorder identification in multi-site MRI imaging: A multi-head cross-attention and multi-context approach for addressing variability in un-harmonized data

Multi-site MRI imaging poses a significant challenge due to the potential variations in images across different scanners at different sites. This variability can introduce ambiguity in further image analysis. Consequently, the image analysis techniques become site-dependent and scanner-dependent, implying that adjustments in the analysis methodologies are necessary for each scanner configuration. Further, implementing real-time modifications becomes intricate, particularly when incorporating a new type of scanner, as it requires adapting the analysis methods accordingly. Taking into account the aforementioned challenge, we have considered its implications for an Autism spectrum disorder (ASD) application. Our objective is to minimize the impact of site and scanner variability in the analysis, aiming to develop a model that remains effective across different scanners and sites. This entails devising a methodology that allows the same model to function seamlessly across multiple scanner configurations and sites. ASD, a behavioral disorder affecting child development, requires early detection. Clinical observation is time-consuming, prompting the use of fMRI with machine/deep learning for expedited diagnosis. Previous methods leverage fMRI's functional connectivity but often rely on less generalized feature extractors and classifiers. Hence, there is significant room for improvement in the generalizability of detection methods across multi-site data, which is acquired from multiple scanners with different settings. In this study, we propose a Cross-Combination Multi-Scale Multi-Context Framework (CCMSMCF) capable of performing neuroimaging-based diagnostic classification of mental disorders for a multi-site dataset. Thus, this framework attains a degree of internal data harmonization, rendering it to some extent site and scanner-agnostic. Our proposed network, CCMSMCF, is constructed by integrating two sub-modules: the Multi-Head Attention Cross-Scale Module (MHACSM) and the Residual Multi-Context Module (RMCN). We also employ multiple loss functions in a novel manner for training the model, which includes Binary Cross Entropy, Dice loss, and Embedding Coupling loss. The model is validated on the Autism Brain Imaging Data Exchange I (ABIDE-I) dataset, which includes data from multiple scanners across different sites, and achieves promising results.

Discriminative possibilistic clustering promoting cross-domain emotion recognition

The affective Brain-Computer Interface (aBCI) systems strive to enhance prediction accuracy for individual subjects by leveraging data from multiple subjects. However, significant differences in EEG (Electroencephalogram) feature patterns among subjects often hinder these systems from achieving the desired outcomes. Although studies have attempted to address this challenge using subject-specific classifier strategies, the scarcity of labeled data remains a major hurdle. In light of this, Domain Adaptation (DA) technology has gradually emerged as a prominent approach in the field of EEG-based emotion recognition, attracting widespread research interest. The crux of DA learning lies in resolving the issue of distribution mismatch between training and testing datasets, which has become a focal point of academic attention. Currently, mainstream DA methods primarily focus on mitigating domain distribution discrepancies by minimizing the Maximum Mean Discrepancy (MMD) or its variants. Nevertheless, the presence of noisy samples in datasets can lead to pronounced shifts in domain means, thereby impairing the adaptive performance of DA methods based on MMD and its variants in practical applications to some extent. Research has revealed that the traditional MMD metric can be transformed into a 1-center clustering problem, and the possibility clustering model is adept at mitigating noise interference during the data clustering process. Consequently, the conventional MMD metric can be further relaxed into a possibilistic clustering model. Therefore, we construct a distributed distance measure with Discriminative Possibilistic Clustering criterion (DPC), which aims to achieve two objectives: (1) ensuring the discriminative effectiveness of domain distribution alignment by finding a shared subspace that minimizes the overall distribution distance between domains while maximizing the semantic distribution distance according to the principle of "sames attract and opposites repel"; and (2) enhancing the robustness of distribution distance measure by introducing a fuzzy entropy regularization term. Theoretical analysis confirms that the proposed DPC is an upper bound of the existing MMD metric under certain conditions. Therefore, the MMD objective can be effectively optimized by minimizing the DPC. Finally, we propose a domain adaptation in Emotion recognition based on DPC (EDPC) that introduces a graph Laplacian matrix to preserve the geometric structural consistency between data within the source and target domains, thereby enhancing label propagation performance. Simultaneously, by maximizing the use of source domain discriminative information to minimize domain discrimination errors, the generalization performance of the DA model is further improved. Comparative experiments on several representative domain adaptation learning methods using multiple EEG datasets (i.e., SEED and SEED-IV) show that, in most cases, the proposed method exhibits better or comparable consistent generalization performance.

Unsupervised and semi-supervised domain adaptation networks considering both global knowledge and prototype-based local class information for Motor Imagery Classification

The non-stationarity of EEG signals results in variability across sessions, impeding model building and data sharing. In this paper, we propose a domain adaptation method called GPL, which simultaneously considers global knowledge and prototype-based local class information to enhance the classification accuracy of motor imagery signals. Depending on the amount of labeled data available in the target domain, the method is implemented in both unsupervised and semi-supervised versions. Specifically, at the global level, we employ the maximum mean difference (MMD) loss to globally constrain the feature space, achieving comprehensive alignment. In the context of class-level operations, we propose two memory banks designed to accommodate class prototypes in each domain and constrain feature embeddings by applying two prototype-based contrastive losses. The source contrastive loss is used to organize source features spatially based on categories, thereby reconciling inter-class and intra-class relationships, while the interactive contrastive loss is employed to facilitate cross-domain information interaction. Simultaneously, in unsupervised scenarios, to mitigate the adverse effects of excessive pseudo-labels, we introduce an entropy-aware strategy that dynamically evaluates the confidence level of target data and personalized constraints on the participation of interactive contrastive loss. To validate our approach, extensive experiments were conducted on a highly regarded public EEG dataset, namely Dataset IIa of the BCI Competition IV, as well as a large-scale EEG dataset called GigaDB. The experiments yielded average classification accuracies of 86.03% and 84.22% respectively. These results demonstrate that our method is an effective EEG decoding model, conducive to advancing the development of motor imagery brain-computer interfaces. The architecture proposed in this study and the code for data partitioning can be found at https://github.com/zhangdx21/GPL.

Locality Cross-domain Discriminant Analysis for Membranous Nephropathy Recognition Using Microscopic Hyperspectral Imaging

Cross-domain methods have been proposed to learn the domain invariant knowledge that can be transferred from the source domain to the target domain. Existing cross-domain methods attempt to minimize the distribution discrepancy of the domains. However, these methods fail to explore the domain invariant subspace due to the samples of different classes between two domains may overlap in the new subspace. They consider the features in the original space data that may be unnecessary or irrelevant to the final classification, and neglect to preserve the local manifold structure between two domains. To solve these problems, a novel feature extraction method called Locality Cross-domain Discriminant Analysis (LCDA) is proposed. LCDA first aligns the distributions and avoids overlap between two domains. Then, LCDA exploits the local manifold structure to maintain the discriminative capability of the low-dimensional projection matrices. Finally, a robust constraint is utilized to preserve the robustness of the projection matrices. The proposed LCDA not only avoids overlap between different classes but also explores the local manifold information. Experiment results on the medical membranous nephropathy hyperspectral dataset demonstrate that the proposed LCDA has better performance than other relevant feature extraction methods.

Supervised Factor Analysis Transfer: Calibration transfer with noise modeling and response variable integration

Multivariate calibration models often encounter challenges in extrapolating beyond the calibration instruments due to variations in hardware configurations, signal processing algorithms, or environmental conditions. Calibration transfer techniques have been developed to mitigate this issue. In this study, we introduce a novel methodology known as Supervised Factor Analysis Transfer (SFAT) aimed at achieving robust and interpretable calibration transfer. SFAT operates from a probabilistic framework and integrates response variables into its transfer process to effectively align data from the target instrument to that of the source instrument. Within the SFAT model, the data from the source instrument, the target instrument, and the response variables are collectively projected onto a shared set of latent variables. These latent variables serve as the conduit for information transfer between the three distinct domains, thereby facilitating effective spectra transfer. Moreover, SFAT explicitly models the noise variances associated with each variable, thereby minimizing the transfer of non-informative noise. Furthermore, we provide empirical evidence showcasing the efficacy of SFAT across three real-world datasets, demonstrating its superior performance in calibration transfer scenarios.

Cross-subject emotion recognition in brain-computer interface based on frequency band attention graph convolutional adversarial neural networks

BACKGROUND

Emotion is an important area in neuroscience. Cross-subject emotion recognition based on electroencephalogram (EEG) data is challenging due to physiological differences between subjects. Domain gap, which refers to the different distributions of EEG data at different subjects, has attracted great attention for cross-subject emotion recognition.

COMPARISON WITH EXISTING METHODS

This study focuses on narrowing the domain gap between subjects through the emotional frequency bands and the relationship information between EEG channels. Emotional frequency band features represent the energy distribution of EEG data in different frequency ranges, while relationship information between EEG channels provides spatial distribution information about EEG data.

NEW METHOD

To achieve this, this paper proposes a model called the Frequency Band Attention Graph convolutional Adversarial neural Network (FBAGAN). This model includes three components: a feature extractor, a classifier, and a discriminator. The feature extractor consists of a layer with a frequency band attention mechanism and a graph convolutional neural network. The mechanism effectively extracts frequency band information by assigning weights and Graph Convolutional Networks can extract relationship information between EEG channels by modeling the graph structure. The discriminator then helps minimize the gap in the frequency information and relationship information between the source and target domains, improving the model's ability to generalize.

RESULTS

The FBAGAN model is extensively tested on the SEED, SEED-IV, and DEAP datasets. The accuracy and standard deviation scores are 88.17% and 4.88, respectively, on the SEED dataset, and 77.35% and 3.72 on the SEED-IV dataset. On the DEAP dataset, the model achieves 69.64% for Arousal and 65.18% for Valence. These results outperform most existing models.

CONCLUSIONS

The experiments indicate that FBAGAN effectively addresses the challenges of transferring EEG channel domain and frequency band domain, leading to improved performance.

Active Dynamic Weighting for multi-domain adaptation

Multi-source unsupervised domain adaptation aims to transfer knowledge from multiple labeled source domains to an unlabeled target domain. Existing methods either seek a mixture of distributions across various domains or combine multiple single-source models for weighted fusion in the decision process, with little insight into the distributional discrepancy between different source domains and the target domain. Considering the discrepancies in global and local feature distributions between different domains and the complexity of obtaining category boundaries across domains, this paper proposes a novel Active Dynamic Weighting (ADW) for multi-source domain adaptation. Specifically, to effectively utilize the locally advantageous features in the source domains, ADW designs a multi-source dynamic adjustment mechanism during the training process to dynamically control the degree of feature alignment between each source and target domain in the training batch. In addition, to ensure the cross-domain categories can be distinguished, ADW devises a dynamic boundary loss to guide the model to focus on the hard samples near the decision boundary, which enhances the clarity of the decision boundary and improves the model's classification ability. Meanwhile, ADW applies active learning to multi-source unsupervised domain adaptation for the first time, guided by dynamic boundary loss, proposes an efficient importance sampling strategy to select target domain hard samples to annotate at a minimal annotation budget, integrates it into the training process, and further refines the domain alignment at the category level. Experiments on various benchmark datasets consistently demonstrate the superiority of our method.

Ada-LT IP: Functional Discriminant Analysis of Feature Extraction for Adaptive Long-Term Wi-Fi Indoor Localization in Evolving Environments

Wi-Fi fingerprint-based indoor localization methods are effective in static environments but encounter challenges in dynamic, real-world scenarios due to evolving fingerprint patterns and feature spaces. This study investigates the temporal variations in signal strength over a 25-month period to enhance adaptive long-term Wi-Fi localization. Key aspects explored include the significance of signal features, the effects of sampling fluctuations, and overall accuracy measured by mean absolute error. Techniques such as mean-based feature selection, principal component analysis (PCA), and functional discriminant analysis (FDA) were employed to analyze signal features. The proposed algorithm, Ada-LT IP, which incorporates data reduction and transfer learning, shows improved accuracy compared to state-of-the-art methods evaluated in the study. Additionally, the study addresses multicollinearity through PCA and covariance analysis, revealing a reduction in computational complexity and enhanced accuracy for the proposed method, thereby providing valuable insights for improving adaptive long-term Wi-Fi indoor localization systems.

Adversarial Training Based Domain Adaptation of Skin Cancer Images

Skin lesion datasets used in the research are highly imbalanced; Generative Adversarial Networks can generate synthetic skin lesion images to solve the class imbalance problem, but it can result in bias and domain shift. Domain shifts in skin lesion datasets can also occur if different instruments or imaging resolutions are used to capture skin lesion images. The deep learning models may not perform well in the presence of bias and domain shift in skin lesion datasets. This work presents a domain adaptation algorithm-based methodology for mitigating the effects of domain shift and bias in skin lesion datasets. Six experiments were performed using two different domain adaptation architectures. The domain adversarial neural network with two gradient reversal layers and VGG13 as a feature extractor achieved the highest accuracy and F1 score of 0.7567 and 0.75, respectively, representing an 18.47% improvement in accuracy over the baseline model.

A Novel Cross-Domain Mechanical Fault Diagnosis Method Fusing Acoustic and Vibration Signals by Vision Transformer

Changes in operating conditions often cause the distribution of signal features to shift during the bearing fault diagnosis process, which will result in reduced diagnostic accuracy of the model. Therefore, this paper proposes a dual-channel parallel adversarial network (DPAN) based on vision transformer, which extracts features from acoustic and vibration signals through parallel networks and enhances feature robustness through adversarial training during the feature fusion process. In addition, the Wasserstein distance is used to reduce domain differences in the fused features, thereby enhancing the network's generalization ability. Two sets of bearing fault diagnosis experiments were conducted to validate the effectiveness of the proposed method. The experimental results show that the proposed method achieves higher diagnostic accuracy compared to other methods. The diagnostic accuracy of the proposed method can exceed 98%.

Cross-Scene Joint Classification of Multisource Data With Multilevel Domain Adaption Network

Domain adaption (DA) is a challenging task that integrates knowledge from source domain (SD) to perform data analysis for target domain. Most of the existing DA approaches only focus on single-source-single-target setting. In contrast, multisource (MS) data collaborative utilization has been extensively used in various applications, while how to integrate DA with MS collaboration still faces great challenges. In this article, we propose a multilevel DA network (MDA-NET) for promoting information collaboration and cross-scene (CS) classification based on hyperspectral image (HSI) and light detection and ranging (LiDAR) data. In this framework, modality-related adapters are built, and then a mutual-aid classifier is used to aggregate all the discriminative information captured from different modalities for boosting CS classification performance. Experimental results on two cross-domain datasets show that the proposed method consistently provides better performance than other state-of-the-art DA approaches.

Riemannian Locality Preserving Method for Transfer Learning With Applications on Brain-Computer Interface

Brain-computer interfaces (BCIs) have been widely focused and extensively studied in recent years for their huge prospect of medical rehabilitation and commercial applications. Transfer learning exploits the information in the source domain and applies in another different but related domain (target domain), and is therefore introduced into the BCIs to figure out the inter-subject variances of electroencephalography (EEG) signals. In this article, a novel transfer learning method is proposed to preserve the Riemannian locality of data structure in both the source and target domains and simultaneously realize the joint distribution adaptation of both domains to enhance the effectiveness of transfer learning. Specifically, a Riemannian graph is first defined and constructed based on the Riemannian distance to represent the Riemannian geometry information. To simultaneously align the marginal and conditional distribution of source and target domains and preserve the Riemannian locality of data structure in both domains, the Riemannian graph is embedded in the joint distribution adaptation (JDA) framework and forms the proposed Riemannian locality preserving-based transfer learning (RLPTL). To validate the effect of the proposed method, it is compared with several existing methods on two open motor imagery datasets, and both multi-source domains (MSD) and single-source domains (SSD) experiments are considered. Experimental results show that the proposed method achieves the highest accuracies in MSD and SSD experiments on three datasets and outperforms eight baseline methods, which demonstrates that the proposed method creates a feasible and efficient way to realize transfer learning.

Unsupervised heterogeneous domain adaptation for EEG classification

Objective.Domain adaptation has been recognized as a potent solution to the challenge of limited training data for electroencephalography (EEG) classification tasks. Existing studies primarily focus on homogeneous environments, however, the heterogeneous properties of EEG data arising from device diversity cannot be overlooked. This motivates the development of heterogeneous domain adaptation methods that can fully exploit the knowledge from an auxiliary heterogeneous domain for EEG classification.Approach.In this article, we propose a novel model named informative representation fusion (IRF) to tackle the problem of unsupervised heterogeneous domain adaptation in the context of EEG data. In IRF, we consider different perspectives of data, i.e. independent identically distributed (iid) and non-iid, to learn different representations. Specifically, from the non-iid perspective, IRF models high-order correlations among data by hypergraphs and develops hypergraph encoders to obtain data representations of each domain. From the non-iid perspective, by applying multi-layer perceptron networks to the source and target domain data, we achieve another type of representation for both domains. Subsequently, an attention mechanism is used to fuse these two types of representations to yield informative features. To learn transferable representations, the maximum mean discrepancy is utilized to align the distributions of the source and target domains based on the fused features.Main results.Experimental results on several real-world datasets demonstrate the effectiveness of the proposed model.Significance.This article handles an EEG classification situation where the source and target EEG data lie in different spaces, and what's more, under an unsupervised learning setting. This situation is practical in the real world but barely studied in the literature. The proposed model achieves high classification accuracy, and this study is important for the commercial applications of EEG-based BCIs.