A5037729915- Advanced Image and Video Retrieval Techniques
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Welding Techniques and Residual Stresses
- Remote-Sensing Image Classification
- Advanced Graph Neural Networks
- Multimodal Machine Learning Applications
- Anomaly Detection Techniques and Applications
- Graph Theory and Algorithms
- Evolutionary Algorithms and Applications
- Metaheuristic Optimization Algorithms Research
- Titanium Alloys Microstructure and Properties
- Network Security and Intrusion Detection
- Robotic Locomotion and Control
- Intermetallics and Advanced Alloy Properties
- Adversarial Robustness in Machine Learning
- Complex Network Analysis Techniques
- Human Pose and Action Recognition
- Machine Learning and Data Classification
- Control and Dynamics of Mobile Robots
- Aerospace Engineering and Energy Systems
- Digital Image Processing Techniques
- Advancements in Photolithography Techniques
- Indoor Air Quality and Microbial Exposure
- Infection Control and Ventilation
Xidian University
2021-2025
Xi'an Jiaotong University
2024
North China University of Science and Technology
2024
Tiangong University
2024
Tianjin Chengjian University
2023
First Research Institute of the Ministry of Public Security
2023
Xi'an Shiyou University
2023
China XD Group (China)
2023
Bridge University
2023
Longdong University
2023
Brain-inspired algorithms have become a new trend in next-generation artificial intelligence. Through research on brain science, the intelligence of remote sensing can be effectively improved. This paper summarizes and analyzes essential properties cognise learning recent advance interpretation. Firstly, this introduces structural composition brain. Then, five represent brain-inspired are studied, including multiscale geometry analysis, compressed sensing, attention mechanism, reinforcement...
Influence maximization is a crucial issue for mining the deep information of social networks, which aims to select seed set from network maximize number influenced nodes. To evaluate influence spread efficiently, existing studies have proposed transformations with lower computational costs replace expensive Monte Carlo simulation process. These alternate transformations, based on prior knowledge, induce different search behaviors similar characteristics various perspectives. Specifically, it...
This study introduces a novel cross-modal spatial-spectral interaction Mamba (CMS <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> I-Mamba) for remote sensing image fusion classification. Unlike convolution-based models focusing on local details and Transformer-based with high computational complexity, CMS I-Mamba efficiently global long-range dependencies in linear complexity manner. First, multispectral (MS) panchromatic (PAN) images...
Breast cancer is the most commonly diagnosed in women. Among all types, triple-negative breast particularly challenging to cure because of its high recurrence rates and invasive metastatic capacity. Although numerous studies have explored role TP53 mutations cancer, there a dearth research regarding correlation between cell proliferation. In this study, our aim was examine impact on prognosis patients with bioinformatics techniques. To detect proliferation, CCK8 assay performed, western...
Solving the complex challenges of sophisticated terrain and multi-scale targets in remote sensing (RS) images requires a synergistic combination Transformer convolutional neural network (CNN). However, crafting effective CNN architectures remains major challenge. To address these difficulties, this study introduces knowledge guided evolutionary for RS scene classification (Evo RSFormer). It amalgamates adaptive CNN) with Transformers hybrid strategy synergistically, which combines...
Fusion and interaction of multimodal features are essential for video question answering. Structural information composed the relationships between different objects in videos is very complex, which restricts understanding reasoning. In this paper, we propose a quaternion hypergraph network (QHGN) answering, to simultaneously involve structural information. Since operations suitable interactions, four components vectors applied represent features. Furthermore, construct based on visual...
The construction of machine learning models involves many bi-level multi-objective optimization problems (BL-MOPs), where upper level (UL) candidate solutions must be evaluated via training weights a model in the lower (LL). Due to Pareto optimality sub-problems and complex dependency across UL LL weights, an solution is feasible if only weight optimal. It computationally expensive determine which set most appropriate for each solution. This paper proposes framework (BLMOL), coupling above...
The foundation model (FM) has garnered significant attention for its remarkable transfer performance in downstream tasks. Typically, it undergoes task-agnosticpre-training on a large dataset and can be efficiently adapted to various applications through fine-tuning. While FMs have been extensively explored language other domains, their potential remote sensing also begun attract scholarly interest. However, comprehensive investigations comparisons of these models tasks are currently lacking....
Deep representation learning has improved automatic remote change detection (RSCD) in recent years. Existing methods emphasize primarily convolutional neural networks (CNNs) or Transformer-based networks. However, most of them neither effectively combine CNNs and Transformer nor use prior geometric information to refine regions. In this paper, a novel (GeoFormer) is proposed for high-resolution RSCD. GeoFormer utilizes guide the by employing knowledge. Specifically, consists three carefully...
Abstract The unsatisfactory lithium‐ion conductivity ( σ ) and limited mechanical strength of polymer solid electrolytes hinder their wide applications in solid‐state lithium metal batteries (SSLMBs). Here, a thin piezoelectric electrolyte integrating electromechanical coupling ferroelectric polarization effects has been designed prepared to achieve long‐term stable cycling SSLMBs. Bi 4 Ti 3 O 12 nanoparticle (BIT NPs) loaded poly(vinylidene fluoride‐trifluoroethylene) (P(VDF‐TrFE))...
Zhicheng Guo, Jiaxuan Zhao, Licheng Jiao, Xu Liu, Lingling Li. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers). 2021.
Existing efforts are dedicated to designing many topologies and graph-aware strategies for the graph Transformer, which greatly improve model's representation capabilities. However, manually determining suitable Transformer architecture a specific dataset or task requires extensive expert knowledge laborious trials. This article proposes an evolutionary search (EGTAS) framework automate construction of strong Transformers. We build comprehensive space with micro-level macro-level designs....
Based on self-righting principle of a free-falling cat, multi-rigid-body dynamic model cat robot is built with the consideration swing legs and relation between turning rate angle got. Mechanical structure designed dynamics simulation carried out to validate theoretical analysis. Simulation results indicate that flip becomes larger increase or angular velocity legs. By contrast, change has more influence robot. Adjusting suitable numerical values, energy consumption motion can be reduced effectively.
Abstract Low‐cost fabric‐based top‐emitting polymer light‐emitting devices (Fa‐TPLEDs) have aroused increasing attention due to their remarkable potential applications in wearable displays. However, it is still challenging realize efficient all‐solution‐processed from bottom electrodes top with large‐scale fabrication. Here, a smooth reflective Ag cathode integrated on fabric by one‐step silver mirror reaction and composite transparent anode of polydimethylsiloxane/silver...
A method for the aerial posture adjustment of a cat robot is proposed by imitating motion falling cats. The process phased into three periods: righting reflex, pitch attitude and landing buffer. kinematic equations phase dynamic buffer are built based on theorem moment momentum classical impact theory. changing length swing angle legs to adjust relationship between got calculation. ground reaction forces, vertical acceleration center mass waist joint torque calculated when front rear collide...
Existing efforts are dedicated to designing many topologies and graph-aware strategies for the graph Transformer, which greatly improve model's representation capabilities. However, manually determining suitable Transformer architecture a specific dataset or task requires extensive expert knowledge laborious trials. This paper proposes an evolutionary search framework (EGTAS) automate construction of strong Transformers. We build comprehensive space with micro-level macro-level designs....
Shortening of telomere length (TL) is correlated with many age-related disorders and a hallmark biological aging. This study used proteome-wide Mendelian randomization to identify the protein biomarkers associated length. Protein quantitative trait loci (pQTL) were derived from two studies, deCODE Health (4907 plasma proteins) UK Biobank Pharma Proteomics Project (2923 proteins). Summary data genome-wide association studies (GWAS) for TL obtained (472,174 cases) GWAS Catalog (418,401 cases)....