Sleep apnea (SA) is a common chronic sleep breathing disorder, which would cause stroke, cognitive decline, cardiovascular disease, or even death. The SA symptoms often manifest as frequent breathing interruptions during sleep and most individuals with sleeping disorders are not aware of the SA events. Using a portable device with single-lead ECG signal is an effective way to help an individual to monitor their sleep conditions at home. However, the SA detection performance of ECG-based methods is still difficult to meet the clinical practice requirement. In this study, we propose an end-to-end spatio-temporal learning-based SA detection method, which consists of multiple spatio-temporal blocks. Each block has the identical architecture with a convolutional neural network (CNN) layer, a max-pooling layer, and a bi-gated recurrent unit (BiGRU) layer. This architecture with repeated spatio-temporal blocks can well capture the morphological spatial feature information as well as the temporal feature information from ECG signals. The proposed SA detection model was evaluated on the publicly available datasets of PhysioNet Apnea-ECG dataset (Apnea-ECG) and University College Dublin Sleep Apnea Database (UCDDB). Extensive experimental results show that our proposed SA model on both Apnea-ECG and UCDDB datasets achieves state-of-the-art results, which are obviously superior to existing ECG-based SA detection methods. It means that our proposed method has the potential to be deployed into a healthcare system to provide a sleep monitoring service, which can screen out SA population with high risk and help to take timely interventions to prevent serious consequences.
Cross-Domain Facial Expression Recognition (CD-FER) aims to develop a facial expression recognition model that can be trained in one domain and deliver consistent performance in another. CD-FER poses a significant challenges due to changes in marginal and class distributions between source and target domains. Existing methods primarily emphasize achieving domain-invariant features through global feature adaptation, often neglecting the potential benefits of transferable local features across different domains. To address this issue, we propose a novel framework for CD-FER that combines reliable global–local representation learning and dynamic label weighting. Our framework incorporates two key modules: the Pseudo-Complementary Label Generation (PCLG) module, which leverages pseudo-labels and complementary labels obtained using a credibility threshold to learn domain-invariant global and local features, and the Label Dynamic Weight Matching (LDWM) module, which assesses the learning difficulty of each category and adaptively assigns corresponding label weights, thereby enhancing the classification performance in the target domain. We evaluate our approach through extensive experiments and analyses on multiple public datasets, including RAF-DB, FER2013, CK+, JAFFE, SFW2.0, and ExpW. The experimental results demonstrate that our proposed model outperforms state-of-the-art methods, with an average accuracy improvement of 3.5% across the five datasets.
8 December 2023 in IEEE Transactions on Network and Service Management
End-to-end (E2E) delay modeling is crucial to network operation and optimization, which is the key enabler for Knowledge-Defined Networking (KDN) and network Digital Twins (DT). Neural network-based methods have been widely applied in this research area and have made significant progress. However, previous work modeled E2E delay with node/link states in the network topology graph, ignoring competitive interactions among E2E flows. To this end, we propose a flow interaction graph construction method. By introducing the flow interaction graph, our proposed method can mine the bandwidth contention information among E2E flows. Meanwhile, with the help of the network context-aware encoding method, we propose a flow interaction graph transformer (FI-Graphormer) model, which can effectively utilize the competitive relationships represented in the flow interaction graph. Experimental results on the publicly available datasets of TnCwD, NSFNET and Geant2 show that FI-Graphormer achieves competitive results, which is superior to the state-of-the-art methods.
16 September 2024 in IEEE Transactions on Cloud Computing
Accurate workload forecasting plays a crucial role in optimizing resource allocation, enhancing performance, and reducing energy consumption in cloud data centers. Deep learning-based methods have emerged as the dominant approach in this field, exhibiting exceptional performance. However, most existing methods lack the ability to quantify confidence, limiting their practical decision-making utility. To address this limitation, we propose a novel denoising diffusion probabilistic model (DDPM)-based method, termed WorkloadDiff, for multivariate probabilistic workload prediction. WorkloadDiff leverages both original and noisy signals from input conditions using a two-path neural network. Additionally, we introduce a multi-scale feature extraction method and an adaptive fusion approach to capture diverse temporal patterns within the workload. To enhance consistency between conditions and predicted values, we incorporate a resampling strategy into the inference of WorkloadDiff. Extensive experiments conducted on four public datasets demonstrate the superior performance of WorkloadDiff over all baseline models, establishing it as a robust tool for resource management in cloud data centers. The source code is available at https://github.com/Ak2714/WorkloadDiff .
Acoustic scene classification (ASC) tries to inference information about the environment using audio segments. The inter-class similarity is a significant issue in ASC as acoustic scenes with different labels may sound quite similar. In this paper, the similarity relations amongst scenes are correlated with the classification error. A class hierarchy construction method by using classification error is then proposed and integrated into a multitask learning framework. The experiments have shown that the proposed multitask learning method improves the performance of ASC. On the TUT Acoustic Scene 2017 dataset, we obtain the ensemble fine-grained accuracy of 81.4%, which is better than the state-of-the-art. By using multitask learning, the basic Convolutional Neural Network (CNN) model can be improved by about 2.0 to 3.5 percent according to different spectrograms. The coarse category accuracies (for two to six super-classes) range from 77.0% to 96.2% by single models. On the revised version of the LITIS Rouen dataset, we achieve the ensemble fine-grained accuracy of 83.9%. The multitask learning models obtain an improvement of 1.6% to 1.8% compared to their basic models. The coarse category accuracies range from 94.9% to 97.9% for two to six super-classes with single models.