A5002618466- Advanced Graph Neural Networks
- Topic Modeling
- Domain Adaptation and Few-Shot Learning
- Graph Theory and Algorithms
- Machine Learning in Materials Science
- Intelligent Tutoring Systems and Adaptive Learning
- AI-based Problem Solving and Planning
- Cell Image Analysis Techniques
- Logic, Reasoning, and Knowledge
- Neural Networks and Applications
- Computational Drug Discovery Methods
- Nanoplatforms for cancer theranostics
- Computer Graphics and Visualization Techniques
- Biosensors and Analytical Detection
- Image and Object Detection Techniques
- Imbalanced Data Classification Techniques
- TiO2 Photocatalysis and Solar Cells
- Genetics, Aging, and Longevity in Model Organisms
- Advanced Battery Technologies Research
- Wine Industry and Tourism
- Bayesian Modeling and Causal Inference
- Opinion Dynamics and Social Influence
- Explainable Artificial Intelligence (XAI)
- Culinary Culture and Tourism
- Advanced Numerical Analysis Techniques
Wuhan University
2015-2024
Hubei Zhongshan Hospital
2023
Renmin Hospital of Wuhan University
2022
Ministry of Education of the People's Republic of China
2015
Urinalysis is attractive in non-invasive early diagnosis of bladder cancer compared with clinical gold standard cystoscopy. However, the trace tumor biomarkers urine and particularly complex environment pose significant challenges for urinalysis. Here, a clinically adoptable urinalysis device that integrates molecular-specificity indium gallium zinc oxide field-effect transistor (IGZO FET) biosensor arrays, control panel, an internet terminal directly analyzing five bladder-tumor-associated...
The growth and proliferation of Li dendrites during repeated cycling has long been a crucial issue that hinders the development secondary Li-metal batteries. Building stable robust solid state electrolyte interphase (SEI) on Li-anode surface is regarded as promising strategy to overcome dendrite issues. In this work, we report simple engineer interface chemistry anodes by using tiny amounts dimethyl sulfate (DMS, C2H6SO4) SEI-forming additive. With preferential reduction DMS, an SEI layer...
Knowledge Graphs typically suffer from incompleteness. A popular approach to knowledge graph completion is infer missing by multihop reasoning over the information found along other paths connecting a pair of entities. However, multi-hop still challenging because process usually experiences multiple semantic issue that relation or an entity has meanings. In order deal with situation, we propose novel Hierarchical Reinforcement Learning framework learn chains Graph automatically. Our inspired...
Recently, multi-hop reasoning over incomplete Knowledge Graphs (KGs) has attracted wide attention due to its desirable interpretability for downstream tasks, such as question answer and knowledge graph completion. Multi-Hop is a typical sequential decision problem, which can be formulated Markov process (MDP). Subsequently, some reinforcement learning (RL) based approaches are proposed proven effective train an agent paths sequentially until reaching the target answer. However, these assume...
Meta-paths are important tools for a wide variety of data mining and network analysis tasks in Heterogeneous Information Networks (HINs), due to their flexibility interpretability capture the complex semantic relation among objects. To date, most HIN still relies on hand-crafting meta-paths, which requires rich domain knowledge that is extremely difficult obtain complex, large-scale, schema-rich HINs. In this work, we present novel framework, Meta-path Discovery with Reinforcement Learning...
Abstract Link prediction is essential for identifying hidden relationships within network data, with significant implications fields such as social analysis and bioinformatics. Traditional methods often overlook potential among common neighbors, limiting their effectiveness in utilizing graph information fully. To address this, we introduce a novel approach, Common Neighbor Completion Information Entropy (IECNC), which enhances model expressiveness by considering logical neighbor...
Answering complex queries over incomplete knowledge graphs (KGs) is a challenging job. Most previous works have focused on learning entity/relation embeddings and simulating first-order logic operators with various neural networks. However, they are bottlenecked by the inability to share world improve logical reasoning, thus resulting in suboptimal performance. In this paper, we propose reasoning schema KG upon large language models (LLMs), containing curriculum-based logical-aware...
Determining the types of neurons within a nervous system plays significant role in analysis brain connectomics and investigation neurological diseases. However, efficiency utilizing anatomical, physiological, or molecular characteristics is relatively low costly. With advancements electron microscopy imaging techniques for tissue, we are able to obtain whole-brain connectome consisting neuronal high-resolution morphology connectivity information. few models built based on such data automated...
Answering complex logical queries over incomplete knowledge graphs (KGs) is challenging. Most previous works have focused on learning entity/relation embeddings and simulating first-order logic operators with various neural networks. However, they are bottlenecked by the inability to share world improve reasoning, thus resulting in suboptimal performance. In this paper, we propose a reasoning schema upon large language models (LLMs), containing curriculum-based logical-aware instruction...
Knowledge Graphs (KGs) provide a structured representation of knowledge but often suffer from challenges incompleteness. To address this, link prediction or graph completion (KGC) aims to infer missing new facts based on existing in KGs. Previous embedding models are limited their ability capture expressive features, especially when compared deeper, multi-layer models. These approaches also assign single static each entity and relation, disregarding the fact that entities relations can...
Abstract Motivation Reconstructing neuron-level brain circuit network is a universally recognized formidable task. A significant impediment involves discerning the intricate interconnections among multitudinous neurons in complex network. However, majority of current methodologies only rely on learning local visual synapse features while neglecting incorporation comprehensive global topological connectivity information. In this paper, we consider perspective and introduce graph neural...
Answering complex queries with First-order logical operators over knowledge graphs, such as conjunction (∧), disjunction (∨), and negation (¬) is immensely useful for identifying missing knowledge. Recently, neural symbolic reasoning methods have been proposed to map entities relations into a continuous real vector space model differential networks. However, traditional methodss employ negative sampling, which corrupts train embeddings. Consequently, these embeddings are susceptible...
Determining the types of neurons within a nervous system plays significant role in analysis brain connectomics and investigation neurological diseases. However, efficiency utilizing anatomical, physiological, or molecular characteristics is relatively low costly. With advancements electron microscopy imaging techniques for tissue, we are able to obtain whole-brain connectome consisting neuronal high-resolution morphology connectivity information. few models built based on such data automated...