A5100462079- Rough Sets and Fuzzy Logic
- Advanced Algebra and Logic
- Multi-Criteria Decision Making
- Logic, Reasoning, and Knowledge
- Fuzzy Logic and Control Systems
- Fuzzy and Soft Set Theory
- Advanced Computational Techniques and Applications
- Natural Language Processing Techniques
- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Multimodal Machine Learning Applications
- Topic Modeling
- Indoor and Outdoor Localization Technologies
- Imbalanced Data Classification Techniques
- Robotic Path Planning Algorithms
- Neural Networks and Applications
- Semantic Web and Ontologies
- Underwater Vehicles and Communication Systems
- Data Mining Algorithms and Applications
- Bayesian Modeling and Causal Inference
- Image and Signal Denoising Methods
- Domain Adaptation and Few-Shot Learning
- Image Enhancement Techniques
- Recommender Systems and Techniques
- Privacy-Preserving Technologies in Data
Zhejiang University
2002-2025
Zhejiang Sci-Tech University
2022-2025
Huazhong University of Science and Technology
2023-2024
Beijing Institute of Technology
2022-2024
State Key Laboratory of Industrial Control Technology
2022-2024
Xi'an University of Architecture and Technology
2024
Northwestern Polytechnical University
2022-2024
South China University of Technology
2022-2024
Southeast University
2024
Yangtze River Pharmaceutical Group (China)
2024
In edge computing (EC), federated learning (FL) enables massive devices to collaboratively train AI models without exposing local data. order avoid the possible bottleneck of parameter server (PS) architecture, we concentrate on decentralized (DFL), which adopts peer-to-peer (P2P) communication maintaining a global model. However, due intrinsic features EC, e.g., resource limitation and heterogeneity, network dynamics non-IID data, DFL with fixed P2P topology and/or an identical model...
Abstract Aims Diabetic cardiomyopathy (DCM) is a major cause of mortality in patients with diabetes, and the potential strategies for treating DCM are insufficient. Melatonin (Mel) has been shown to attenuate DCM, however, underlying mechanism remains unclear. The role vascular endothelial growth factor-B (VEGF-B) little known. In present study, we aimed investigate whether Mel alleviated via regulation VEGF-B explored its mechanisms. Methods results We found that significantly cardiac...
Moral norms vary across cultures. A recent line of work suggests that English large language models contain human-like moral biases, but these studies typically do not examine variation in a diverse cultural setting. We investigate the extent to which monolingual knowledge about different countries. consider two levels analysis: 1) whether capture fine-grained countries over variety topics such as "homosexuality" and "divorce"; 2) diversity shared tendencies people around globe tend diverge...
Abstract The optoelectronic resistive random‐access memory (RRAM) with the integrated function of perception, storage and intrinsic randomness displays promising applications in hardware level in‐sensor image cryptography. In this work, 2D hexagonal boron nitride based RRAM is fabricated semitransparent noble metal (Ag or Au) as top electrodes, which can simultaneous capture color generate physically unclonable (PUF) key for Surface plasmons metals enable strong light absorption to realize...
This article focuses on feature-based underwater localization and navigation for autonomous vehicles (AUVs) using 2D imaging sonar measurements. The sparsity of acoustic features the loss elevation angle in images may introduce wrong feature matches or insufficient optimization-based (i.e. under-constrained/degeneracy cases). motivates us to propose a novel inertial-aided sliding window optimization framework improve estimation accuracy robustness front-end outliers. Concretely, we first...
Abstract Tumour segmentation in medical images (especially 3D tumour segmentation) is highly challenging due to the possible similarity between tumours and adjacent tissues, occurrence of multiple variable shapes sizes. The popular deep learning‐based algorithms generally rely on convolutional neural network (CNN) Transformer. former cannot extract global image features effectively while latter lacks inductive bias involves complicated computation for volume data. existing hybrid...
As the power demand of computing cores grows explosively, large conversion ratio, high efficiency and compact 12V-to-1V DC-DC converters become increasingly important. Double step-down (DSD) [1]–[2], also known as series-capacitor converters, offer a promising solution using capacitors to reduce voltage stress extend duty cycles, however, at cost multiple inductors limited cycle range (D <tex xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Information-based autonomous robot exploration methods, aiming to maximize the rewards, e.g., mutual information (MI), get more prevalent in field robotics applications. However, most MI-based methods assume known poses or use inaccurate pose uncertainty approximation, which may lead deviation even failure when exploring prior unknown environments. In this paper, we explicitly consider full-state (pose & map) for balancing and localizability, i.e., avoiding guiding itself complex scenes with...
Accurate forecasting of photovoltaic (PV) power is essential for grid scheduling and energy management. In recent years, deep learning technology has made significant progress in time-series forecasting, offering new solutions PV forecasting. This study provides a systematic review models concentrating on comparisons the features, advantages, limitations different model architectures. First, we analyze commonly used datasets Additionally, provide an overview mainstream architectures,...
Anti-icing/de-icing coatings on equipment surfaces are becoming an important protective measure against frost formation and buildup, especially when they used in very harsh environments. Herein, perfluorodecyltriethoxysilane (FAS) has been grafted onto the surface of graphene (GPE) via hydrolysis alkaline condition. The FAS modified (F-GPE) is hydrophobic conductive. A photothermal electrothermal superhydrophobic coating obtained by mixing F-GPE into epoxy resin as a filler for anti/de-icing...
Conventional methods that merge multiple images with different exposure levels often suffer from blur and ghosting due to object movement. Existing removal algorithms are usually complex slow, making them unsuitable for real-time video applications. To address this challenge, study proposes a novel ghosting-free image fusion method using dual gain stream setup on an FPGA. IMX662 sensor is employed, which simultaneously captures both HCG LCG the same time, enabling efficient HDR synthesis....
Promoting the digital transformation of education is a crucial move in implementing strategies rejuvenating country through science and education, building strong nation talent, driving development innovation. The essence educational model for innovation entrepreneurship integration 'theory + practice.' However, practical teaching has consistently been weak point various universities, lacking guidance from enterprises high-quality case studies. Therefore, it necessary to tailor content based...
Differential evolution (DE) algorithm is recognized as one of the most effective evolutionary algorithms, demonstrating remarkable efficacy in black-box optimization due to its derivative-free nature. Numerous enhancements fundamental DE have been proposed, incorporating innovative mutation strategies and sophisticated parameter tuning techniques improve performance. However, no single variant has proven universally superior across all problems. To address this challenge, we introduce a...