Explore the vast range of titles available.

Most logit-based knowledge distillation methods transfer soft labels from the teacher model to the student model via Kullback–Leibler divergence based on softmax, an exponential normalization function. However, this exponential nature of softmax tends to prioritize the largest class (target class) while neglecting smaller ones (non-target classes), leading to an oversight of the non-target classes’s significance. To address this issue, we propose Non-Target-Class-Enhanced Knowledge Distillation (NTCE-KD) to amplify the role of non-target classes both in terms of magnitude and diversity. Specifically, we present a magnitude-enhanced Kullback–Leibler (MKL) divergence multi-shrinking the target class to enhance the impact of non-target classes in terms of magnitude. Additionally, to enrich the diversity of non-target classes, we introduce a diversity-based data augmentation strategy (DDA), further enhancing overall performance. Extensive experimental results on the CIFAR-100 and ImageNet-1k datasets demonstrate that non-target classes are of great significance and that our method achieves state-of-the-art performance across a wide range of teacher–student pairs.

In this paper, we propose a novel semi-online approach to tracking multiple people. In contrast to conventional offline approaches that take the whole image sequence as input, our semi-online approach tracks people in a frame-by-frame manner by exploring the time, space and multi-camera relationship of detection hypotheses in the near future frames. We cast the multi-people tracking task as a re-identification problem, and explicitly account for objects’ appearance changes and longer-term associations. We model our approach using a Multi-Label Markov Random Field, and introduce a fast α-expansion algorithm to solve it efficiently. To our best knowledge, this is the first semi-online approach achieved by re-identification. It yields very promising tracking results especially in challenging cases, such as scenarios of the crowded streets where pedestrians frequently occlude each other, scenes captured with moving cameras where objects may disappear and reappear randomly, and videos under changing illuminations wherein the appearances of objects are influenced.

Remote sensing image (RSI) scene classification aims to identify semantic categories in RSI using neural networks. However, high-performance deep neural networks typically demand substantial storage and computational resources, making practical deployment challenging. Knowledge distillation has emerged as an effective technique for developing compact models that maintain high classification accuracy in RSI tasks. Existing knowledge distillation methods often overlook the high inter-class similarity in RSI scenes, leading to low-confidence soft labels from the teacher model, which can mislead the student model. Conversely, overly confident soft labels may discard valuable non-target information. Additionally, the significant intra-class variability in RSI contributes to instability in the model’s decision boundaries. To address these challenges, we propose an efficient method called instance-level scaling and dynamic margin-alignment knowledge distillation (ISDM) for RSI scene classification. To balance the target and non-target class influence, we apply an entropy regularization loss to scale the teacher model’s target class at the instance level. Moreover, we introduce dynamic margin alignment between the student and teacher models to improve the student’s discriminative capability. By optimizing soft labels and enhancing the student’s ability to distinguish between classes, our method reduces the effects of inter-class similarity and intra-class variability. Experimental results on three public RSI scene classification datasets (AID, UCMerced, and NWPU-RESISC) demonstrate that our method achieves state-of-the-art performance across all teacher–student pairs with lower computational costs. Additionally, we validate the generalization of our approach on general datasets, including CIFAR-100 and ImageNet-1k.

The aim of multi-agent reinforcement learning systems is to provide interacting agents with the ability to collaboratively learn and adapt to the behavior of other agents. Typically, an agent receives its private observations providing a partial view of the true state of the environment. However, in realistic settings, the harsh environment might cause one or more agents to show arbitrarily faulty or malicious behavior, which may suffice to allow the current coordination mechanisms fail. In this paper, we study a practical scenario of multi-agent reinforcement learning systems considering the security issues in the presence of agents with arbitrarily faulty or malicious behavior. The previous state-of-the-art work that coped with extremely noisy environments was designed on the basis that the noise intensity in the environment was known in advance. However, when the noise intensity changes, the existing method has to adjust the configuration of the model to learn in new environments, which limits the practical applications. To overcome these difficulties, we present an Attention-based Fault-Tolerant (FT-Attn) model, which can select not only correct, but also relevant information for each agent at every time step in noisy environments. The multihead attention mechanism enables the agents to learn effective communication policies through experience concurrent with the action policies. Empirical results showed that FT-Attn beats previous state-of-the-art methods in some extremely noisy environments in both cooperative and competitive scenarios, much closer to the upper-bound performance. Furthermore, FT-Attn maintains a more general fault tolerance ability and does not rely on the prior knowledge about the noise intensity of the environment.

Recent studies have revealed the critical role of AQP4 in the occurrence and development of gliomas. However, the role of AQP4 in immune regulation has not yet been reported. Many recent reports have identified the lymphatic system's occurrence within the central nervous system (CNS) and the vital role of immune regulation in treating brain tumors. Therefore, the present study aimed to explore the role of AQP4 in the immune regulation of glioma. We used bioinformatics analysis to investigate the immunoregulatory function of AQP4, including its correlation with immunity, anti-tumor immune processes, immunotherapy, immune infiltration, TMB, stemness, mutation, and pan-cancer. The results revealed that AQP4 was significantly associated with the expression of multiple immune checkpoints, immune cells, as well as multiple immune cell effector genes, and antigen presentation and processing abilities. Although no significant correlation was found between the AQP4 gene and IDH mutation and MGMT, AQP4 demonstrated substantial expression differences in different immunophenotypes and molecular types. Using the TTD database, we discovered that EGFR, ABAT, and PDGFRA are strongly associated with AQP4 expression in the glioblastoma (GBM) classification, and these factors could be the potential AQP4-related immunotherapy targets. Afterwards, we screened the differential genes in the high and low AQP4 gene expression group, the high and low immune score group, and the high and low matrix score group and took the intersection as the candidate factor. Finally, univariate Cox analysis was used to find 8 prognostic variables with significant differences across the candidate genes. After lasso dimensionality reduction, three genes built the model (RARRES1, SOCS3, and TTYH1). The scoring model generated by the three genes was eventually obtained after the multi-factor screening of the three genes. Finally, combined with clinical information and cox regression analysis, it was further confirmed that the model score could be used as an independent prognostic factor.

Microglia are the resident immune cells of the CNS, maintaining brain homeostasis partially through phagocytosis. In Alzheimer’s disease (AD), microglial phagocytosis is significantly impaired, contributing to the accumulation of pathological aggregates. Microglial phenotypes are dynamic and can shift depending on the disease stage and local environment. While some subpopulations retain or enhance phagocytic activity, especially under inflammatory conditions, others lose their capacity to clear toxic debris effectively. This variability underscores the need for a more nuanced understanding of microglial regulation and function. This paper explores the dual role of microglial phagocytosis in AD and discusses the emerging insights into microglial heterogeneity and how phenotypic shifts affect phagocytic capacity throughout disease progression. A comprehensive understanding of microglial phagocytosis and its dysregulation in AD is essential for designing targeted treatments. Modulating microglial activity to enhance their protective roles without triggering harmful inflammation represents a promising direction for therapeutic intervention in AD.

Brain tumor (BT) is one of many malignancies that have substantially enhanced global human morbidity and mortality rates. Early detection and characterization of glioma are essential for effective preventive strategies. Currently, the use of Transformers, a deep learning model for BT diagnosis and treatment, is attracting significant attention. The transformer self‐attention mechanism automatically learns the associations between input data for efficient processing and analysis. Research indicates that Transformers could play an essential role in the BT segmentation of magnetic resonance imaging (MRI) images, the MRI and histopathology‐based grading of brain cancer, BT molecular expression prediction, the classification of primary brain metastasis sites, voxel‐level dose and BT radiotherapy outcome prediction, synergistic prediction, and the pathway deconvolution of drug combinations. In this review, the feasibility, accuracy, and applicability of various algorithms are systematically analyzed and their prospects are discussed. Overall, this review aimed to discuss and provide an overview of the increasing applications of Transformers in real‐time BT detection and therapy, indicating their broad prospects and potential. In the future, Transformers are expected to be increasingly used for the diagnosis and subsequent treatment of BT because of the continuous development and improvement of Transformer‐based deep learning technology. However, more work is required to investigate their properties for anomaly detection, medical image classification, network design development, and application to other medical data. Transformers play an essential role in the brain tumor segmentation of MRI images, the MRI‐ and histopathology‐based grading of brain cancer, brain tumor molecular expression prediction, the classification of primary brain metastasis sites, voxel‐level dose and brain tumor radiotherapy outcome prediction, synergistic prediction, and the pathway deconvolution of drug combinations. This review aimed to discuss the increasing applications of Transformers in real‐time brain tumor detection and therapy, indicating their broad prospects and potential.

Cancer of the central nervous system (CNS) can crosstalk systemically and locally in the tumor microenvironment and has become a topic of attention for tumor initiation and advancement. Recently studied neuronal and cancer interaction fundamentally altered the knowledge about glioma and metastases, indicating how cancers invade complex neuronal networks. This review systematically discussed the interactions between neurons and cancers and elucidates new therapeutic avenues. We have overviewed the current understanding of direct or indirect communications of neuronal cells with cancer and the mechanisms associated with cancer invasion. Besides, tumor‐associated neuronal dysfunction and the influence of cancer therapies on the CNS are highlighted. Furthermore, interactions between peripheral nervous system and various cancers have also been discussed separately. Intriguingly and importantly, it cannot be ignored that exosomes could mediate the “wireless communications” between nervous system and cancer. Finally, promising future strategies targeting neuronal–brain tumor interactions were reviewed. A great deal of work remains to be done to elucidate the neuroscience of cancer, and future more research should be directed toward clarifying the precise mechanisms of cancer neuroscience, which hold enormous promise to improve outcomes for a wide range of malignancies. Neurons and cancer cells’ synaptic communication can modulate cancer growth via neurotransmitters and voltage‐mediated mechanisms. Besides, paracrine signaling between cancer and nerve cells, for instance, the neuron‐mediated secretion of growth factors or neurotransmitters, modulates cancer growth in various tissues. The neuronal influence on malignant cells might be direct or affect other cells in TME. Cancer‐induced paracrine factors modulate nervous system to enhance neural activity in TME. Furthermore, cancers functionally control neural networks, and aberrant neural circuits stimulate tumor progression. Circulating factors released from cancer effects nervous system activity, whereas nervous system influences cancer growth by circulating molecules (hormones and progenitor cells) and alters immune system function.

Stock trend prediction refers to predicting future price trend of stocks for seeking profit maximum of stock investment. Although it has aroused broad attention in stock markets, it is still a tough task not only because the stock markets are complex and easily volatile but also because real short-term stock data is so limited that existing stock prediction models could be far from perfect, especially for deep neural networks. As a kind of time-series data, the underlying patterns of stock data are easily influenced by any tiny noises. Thus, how to augment limited stock price data is an open problem in stock trend prediction, since most data augmentation schemes adopted in image processing cannot be brutally used here. To this end, we devise a simple yet effective time-sensitive data augmentation method for stock trend prediction. To be specific, we augment data by corrupting high-frequency patterns of original stock price data as well as preserving low-frequency ones in the frame of wavelet transformation. The proposed method is motivated by the fact that low-frequency patterns without noisy corruptions do not hurt the true patterns of stock price data. Besides, a transformation technique is proposed to recognize the importance of the patterns at varied time points, that is, the information is time-sensitive. A series of experiments carried out on a real stock price dataset including 50 corporation stocks verify the efficacy of our data augmentation method.

Aquaporin 4 (AQP4) is a primary water channel found on astrocytes in the central nervous system (CNS). Besides its function in water and ion homeostasis, AQP4 has also been documented to be involved in a myriad of acute and chronic cerebral pathologies, including autoimmune neurodegenerative diseases. AQP4 has been postulated to be associated with the incidence of a progressive neurodegenerative disorder known as amyotrophic lateral sclerosis (ALS), a disease that targets the motor neurons, causing muscle weakness and eventually paralysis. Raised AQP4 levels were noted in association with vessels surrounded with swollen astrocytic processes as well as in the brainstem, cortex, and gray matter in patients with terminal ALS. AQP4 depolarization may lead to motor neuron degeneration in ALS via GLT-1. Besides, alterations in AQP4 expression in ALS may result in the loss of blood–brain barrier (BBB) integrity. Changes in AQP4 function may also disrupt K+ homeostasis and cause connexin dysregulation, the latter of which is associated to ALS disease progression. Furthermore, AQP4 suppression augments recovery in motor function in ALS, a phenomenon thought to be associated to NGF. No therapeutic drug targeting AQP4 has been developed to date. Nevertheless, the plethora of suggestive experimental results underscores the significance of further exploration into this area.