With the proliferation of wearable devices and ubiquitous computing, context-aware music recommendation systems are evolving to deliver more personalized experiences. Traditional methods relying on explicit feedback, such as listening history and song ratings, struggle to adapt to users' dynamic contextual states, limiting their effectiveness. In this paper, we introduce Wear2Rec, a privacy-preserving, IoT-driven music recommendation system that leverages passive physiological, psychological, and environmental data from wearable devices to enhance personalization. At its core, Wear2Rec employs an innovative dual-expert-dual-task network architecture that separately extracts context and music features, minimizing cross-modal interference. Unlike conventional models, it simultaneously optimizes both music recommendation and mood improvement prediction, ensuring both relevant music suggestions and an emotionally supportive listening experience. Experimental results show its superiority, achieving 0.8411 AUC for music recommendation and 0.5928 MAE for mood prediction, outperforming traditional models by integrating emotional adaptation. Wear2Rec represents a significant step forward in human-centric, real-time recommendation systems, setting new standards for personalized music experiences in IoT-driven ubiquitous computing environments.

Convolutional neural networks (CNNs) often exhibit overfitting due to overconfident predictions, which limits the effective utilization of training samples. Inspired by the diverse effects of training from different samples, we propose selective output smoothing regularization(SOSR) that improves model performance by encouraging the generation of equal logits on incorrect classes when handling samples that are correctly and overconfidently classified. This plug-and-play approach integrates seamlessly into diverse CNN architectures without altering their core design. SOSR demonstrates consistent improvements on various benchmarks, such as a 1.1% accuracy gain on ImageNet with ResNet-50 (77.30%). It synergizes effectively with several widely used techniques, such as CutMix and label smoothing, achieving incremental benefits, highlighting its potential as a foundational tool in advancing deep learning applications. Overall, SOSR effectively alleviates underutilization of high-confidence samples, enhances the generalizability of CNNs, and emerges as a robust tool for improving deep learning applications.

Cross-lingual representation learning transfers knowledge from resource-rich data to resource-scarce ones to improve the semantic understanding abilities of different languages. However, previous works rely on shallow unsupervised data generated by token surface matching, regardless of the global context-aware semantics of the surrounding text tokens. In this paper, we propose an Unsupervised Pseudo Semantic Data Augmentation (UniPSDA) mechanism for cross-lingual natural language understanding to enrich the training data without human interventions. Specifically, to retrieve the tokens with similar meanings for the semantic data augmentation across different languages, we propose a sequential clustering process in 3 stages: within a single language, across multiple languages of a language family, and across languages from multiple language families. Meanwhile, considering the multi-lingual knowledge infusion with context-aware semantics while alleviating computation burden, we directly replace the key constituents of the sentences with the above-learned multi-lingual family knowledge, viewed as pseudo-semantic. The infusion process is further optimized via three de-biasing techniques without introducing any neural parameters. Extensive experiments demonstrate that our model consistently improves the performance on general zero-shot cross-lingual natural language understanding tasks, including sequence classification, information extraction, and question answering.

The supercapacitor tram charges its supercapacitors as passengers enter and exit each station, taking advantage of the short parking time, usually within 30 seconds. However, using a single DC-DC converter for energy conversion would be costly and unreliable. To address this issue, this paper proposes a method that involves the parallel connection of multiple DC-DC converters in coordination with supercapacitors on supercapacitor trams to achieve power balancing. The paper begins by developing a system model for the power supply system and explaining its motivation. Next, a design for a parallel connection of multiple DC-DC converters is presented to prevent excessive current in any one converter and achieve current balancing among them, thereby reducing the system load. Finally, a cooperative control method is proposed to achieve current balancing at each DC-DC converter. The proposed method’s effectiveness has been demonstrated through extensive experiments and simulations.

In electric vehicles equipped with hybrid energy storage systems, it is crucial to achieve fast and precise regulation of the DC bus voltage to match the target reference. Existing techniques often struggle to avoid voltage overshoots while pursuing fast stabilization, which leads to significant fluctuations in the DC bus voltage. To overcome this challenge, this paper presents an advanced energy management strategy using predictive setpoint modulation. We develop an innovative set-point modulation technique that achieves efficient regulation of the DC bus voltage by utilizing a feed-forward compensator in combination with parallel operation of multiple DC-DC converters and supercapacitors (SCs). This configuration not only ensures voltage stability, but also optimizes the charge/discharge cycle management of the supercapacitors, which improves the energy utilization efficiency and overall system performance. Laboratory experimental results show that our strategy significantly reduces DC bus voltage fluctuations and maintains stable system operation under various load conditions, outperforming conventional technologies. In addition, our technology has the advantages of fast response and easy integration into existing systems, providing an efficient and reliable solution for energy management in electric vehicles.

In recent years, various loss functions have been proposed to boost the performance of deep neural networks. Every loss function has its own specific theoretical motivation, and can easily learn its preference features of training data compared with other loss functions. Thus, combining multiple loss functions to capture more data features becomes an attractive idea for model performance improvement. In this paper, instead of using a single loss function or a linear weighted sum of multiple loss functions, we present the method named Multiple Independent Losses Scheduling (MILS), which allows multiple loss functions to independently participate in the training process according to their performance. Specifically, for all candidate loss functions, one loss function will be predefined as the primary loss function before training, and the other loss functions will play auxiliary roles for possible contributions to improve the model performance. In order to avoid auxiliary loss functions bringing a negative effect on the model performance in the training process, we developed a simple but effective performance-based scheduling algorithm to prevent auxiliary loss functions from dragging down the model performance. Extensive experiments using various deep architectures on various recognition benchmarks demonstrate our scheme is simple, robust, lightweight, and effective for typical classification tasks.

Generative Adversarial Networks(GANs) are powerful generative models on numerous tasks and datasets but are also known for their training instability and mode collapse. The latter is because the optimal transportation map is discontinuous, but DNNs can only approximate continuous ones. One way to solve the problem is to introduce multiple discriminators or generators. However, their impacts are limited because the cost function of each component is the same. That is, they are homogeneous. In contrast, multiple discriminators with different cost functions can yield various gradients for the generator, which indicates we can use them to search for more transportation maps in the latent space. Inspired by this, we have proposed a framework to combat the mode collapse problem, containing multiple discriminators with different cost functions, named CES-GAN. Unfortunately, it may also lead to the generator being hard to train because the performance between discriminators is unbalanced, according to the Cannikin Law. Thus, a gradient selecting mechanism is also proposed to pick up proper gradients. We provide mathematical statements to prove our assumptions and conduct extensive experiments to verify the performance. The results show that CES-GAN is lightweight and more effective for fighting against the mode collapse problem than similar works.

A novel coronavirus disease (COVID-19) is a pandemic disease has caused 4 million deaths and more than 200 million infections worldwide (as of August 4, 2021). Rapid and accurate diagnosis of COVID-19 infection is critical to controlling the spread of the epidemic. In order to quickly and efficiently detect COVID-19 and reduce the threat of COVID-19 to human survival, we have firstly proposed a detection framework based on reinforcement learning for COVID-19 diagnosis, which constructs a mixed loss function that can integrate the advantages of multiple loss functions. This paper uses the accuracy of the validation set as the reward value, and obtains the initial model for the next epoch by searching the model corresponding to the maximum reward value in each epoch. We also have proposed a prediction framework that integrates multiple detection frameworks using parameter sharing to predict the progression of patients' disease without additional training. This paper also constructed a higher-quality version of the CT image dataset containing 247 cases screened by professional physicians, and obtained more excellent results on this dataset. Meanwhile, we used the other two COVID-19 datasets as external verifications, and still achieved a high accuracy rate without additional training. Finally, the experimental results show that our classification accuracy can reach 98.31%, and the precision, sensitivity, specificity, and AUC (Area Under Curve) are 98.82%, 97.99%, 98.67%, and 0.989, respectively. The accuracy of external verification can reach 93.34% and 91.05%. What's more, the accuracy of our prediction framework is 91.54%. A large number of experiments demonstrate that our proposed method is effective and robust for COVID-19 detection and prediction.

In this paper, we propose a novel data augmentation strategy named Cut-Thumbnail, that aims to improve the shape bias of the network. We reduce an image to a certain size and replace the random region of the original image with the reduced image. The generated image not only retains most of the original image information but also has global information in the reduced image. We call the reduced image as thumbnail. Furthermore, we find that the idea of thumbnail can be perfectly integrated with Mixed Sample Data Augmentation, so we put one image's thumbnail on another image while the ground truth labels are also mixed, making great achievements on various computer vision tasks. Extensive experiments show that Cut-Thumbnail works better than state-of-the-art augmentation strategies across classification, fine-grained image classification, and object detection. On ImageNet classification, ResNet-50 architecture with our method achieves 79.21% accuracy, which is more than 2.8% improvement on the baseline.