A5100323217- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Advanced Graph Neural Networks
- Recommender Systems and Techniques
- Domain Adaptation and Few-Shot Learning
- Frequency Control in Power Systems
- Multimodal Machine Learning Applications
- Service-Oriented Architecture and Web Services
- IoT and Edge/Fog Computing
- Semantic Web and Ontologies
- Advanced optical system design
- Machine Learning and Data Classification
- Quantum Information and Cryptography
- Blockchain Technology Applications and Security
- Explainable Artificial Intelligence (XAI)
- Computational Geometry and Mesh Generation
- Biomedical Text Mining and Ontologies
- Machine Learning in Materials Science
- Parallel Computing and Optimization Techniques
- Medical Image Segmentation Techniques
- Cold Atom Physics and Bose-Einstein Condensates
- Robotics and Sensor-Based Localization
- Gamma-ray bursts and supernovae
- Automated Road and Building Extraction
- Age of Information Optimization
Zhejiang Lab
2023-2025
Chinese Academy of Sciences
2015-2023
Institute of Computing Technology
2022-2023
Nanjing Institute of Geology and Paleontology
2022
Institute of High Energy Physics
2022
Changchun Institute of Optics, Fine Mechanics and Physics
2021
Shandong Institute of Automation
2015
Shenzhen Institutes of Advanced Technology
2014
University of Science and Technology of China
2004
Despite the progress made by few-shot segmentation (FSS) in low-data regimes, generalization capability of most previous works could be fragile when countering hard query samples with seen-class objects. This paper proposes a fresh and powerful scheme to tackle such an intractable bias problem, dubbed base meta (BAM). Concretely, we apply auxiliary branch (base learner) conventional FSS framework (meta explicitly identify base-class objects, i.e., regions that do not need segmented. Then,...
ABSTRACT Gravitational wave high-energy Electromagnetic Counterpart All-sky Monitor (GECAM) is a dedicated mission, launched in December 2020, for gamma-ray transients, including bursts (GRBs) and soft gamma repeater (SGR) the multimessenger multiwavelength astronomy era. Since GECAM consists of two independent spacecrafts (or say instruments), framework data analysis multiple distinctive from that only one spacecraft, which case most GRB missions, we developed pipeline called Energetic...
Recently, recommender systems based on knowledge graphs (KGs) have become a popular research direction. Graph neural network (GNN) is the key technology of KG-based recommendation systems. However, existing GNNs significant flaw: They cannot explicitly model users' intent in recommendations. Intent plays an essential role behaviors. For example, users may first generate to purchase certain group items and then select specific item from their preferences. Therefore, modeling has positive...
Edge computing has become increasingly important to fulfill the diversified Quality-of-Service (QoS) or Quality-of-Experience (QoE) demands for Industrial Internet of Things (IIoT) applications, such as machine condition monitoring, fault diagnosis, intelligent production scheduling, and quality control. Due heterogeneity IIoT systems, it is urgent necessity concentrate on cloud–edge–end cooperative partial task offloading resource allocation (CPTORA) problem realizing workload balancing,...
As one of the most-critical components in space optical cameras, performance mirrors directly affects imaging quality and lightweight form mirror blanks is a key factor affecting structural surface-shape accuracy mirrors. For design requirements high with mirrors, this study proposes manufacturing method that integrates topology-optimization additive-manufacturing technology. This article firstly introduced basic process technologies space-mirror analyzed superiority combining...
Stratigraphic knowledge, the cornerstone of geoscience, needs to be represented by Knowledge Graph based upon ontology, in order apply state-of-the-art big-data techniques. This study aims comprehensively construct ontologies for stratigraphic domain. has been achieved a federated, crowd intelligence-based collaboration among domain experts major subdisciplines. The initial step is enumerate key terms from authoritative references and incorporate them into Geoscience Professional Graphs...
Click-through rate (CTR) estimation plays as a pivotal function module in various online services. Previous studies mainly apply CTR models to the field of recommendation or advertisement. Indeed, is also critical information retrieval, since probability can serve valuable feature for query-document pair. In this paper, we study task under account search scenario WeChat, where users official accounts mini programs corresponding an organization. Despite large number models, directly applying...
For the two-mode fermionic squeezing operators we find that their coherent state projection operator representation make up a loyal representation, which is homomorphic to an SO(4) group, though states are not mutual orthogonal. Thus result of successively operating with many on can be equivalent single operation. The mappings orthogonal transformations in Grassmann number pseudo-classical space representation.
Differentiable architecture search (DARTS) has attracted much attention due to its simplicity and significant improvement in efficiency. However, the excessive accumulation of skip connection makes it suffer from long-term weak stability low robustness. Many works attempt restrict connections by indicators or manual design, however, these methods are susceptible thresholds human priors. In this work, we suggest a more subtle direct approach that removes operation space. Then, introducing an...