A5086903789- Machine Learning and ELM
- Machine Learning and Data Classification
- Advanced Computational Techniques and Applications
- Computational Drug Discovery Methods
- Advanced Algorithms and Applications
- Machine Learning in Materials Science
- Iron and Steelmaking Processes
- Rough Sets and Fuzzy Logic
- Machine Learning in Healthcare
- Privacy-Preserving Technologies in Data
- AI in cancer detection
- Viral Infections and Outbreaks Research
- Geographic Information Systems Studies
- Anomaly Detection Techniques and Applications
- Disaster Response and Management
- Recommender Systems and Techniques
- Image Processing and 3D Reconstruction
- Radiative Heat Transfer Studies
- IoT and Edge/Fog Computing
- Cell Image Analysis Techniques
- Topic Modeling
- Microbial Natural Products and Biosynthesis
- Statistical Methods and Inference
- Artificial Intelligence in Healthcare
- Text and Document Classification Technologies
University of Jinan
2024
Chinese Academy of Medical Sciences & Peking Union Medical College
2022-2024
Kunming University of Science and Technology
2013-2024
Harbin Electric Corporation (China)
2022-2023
Northeastern University
2022
Chinese Center For Disease Control and Prevention
2016
Anshan Normal University
2011
Wuhan University
2009
Chinese Academy of Surveying and Mapping
2009
China Academy of Information and Communications Technology
2009
Discovery of small-molecule antibiotics with novel chemotypes serves as one the essential strategies to address antibiotic resistance. Although a considerable number computational tools committed molecular design have been reported, there is deficit in holistic and efficient specifically developed for discovery. To this issue, we report AutoMolDesigner, modeling software dedicated design. It generalized framework comprising two functional modules, i.e., generative-deep-learning-enabled...
Low-Rank Adaptation (LoRA) is widely used for adapting large language models (LLMs) to specific domains due its efficiency and modularity. Meanwhile, vanilla LoRA struggles with task conflicts in multi-task scenarios. Recent works adopt Mixture of Experts (MoE) by treating each module as an expert, thereby mitigating interference through multiple specialized modules. While effective, these methods often isolate knowledge within individual tasks, failing fully exploit the shared across...
Abstract Multi-Object Tracking (MOT) is an important topic in computer vision. Recent MOT methods based on the anchor-free paradigm trade complicated hierarchical structures for tracking performance. However, existing ignore noise detection, data association, and trajectory reconnection stages, which results serious problems, such as missing detection of small objects, insufficient motion information, drifting. To solve these this paper proposes Noise-Control Tracker (NCT), focuses...
Celastrol (CEL), a pentacyclic triterpene compound, has been proven to have definite antipulmonary fibrosis effect. However, its direct targets for remain unknown. In this study, we designed and synthesized series of celastrol-based probes identify the in human pulmonary fibroblasts using an activity-based protein profiling strategy. Among many fished targets, identified key protein, cullin-associated neddylation-dissociated 1 (CAND1), which was involved fibroblast–myofibroblast...
Graph neural networks (GNN) have been successful in many fields, and derived various researches applications real industries. However, some privacy sensitive scenarios (like finance, healthcare), training a GNN model centrally faces challenges due to the distributed data silos. Federated learning (FL) is an emerging technique that can collaboratively train shared while keeping decentralized, which rational solution for training. We term it as federated graph (FGL). Although FGL has received...
Low-Rank Adaptation (LoRA) offers an efficient way to fine-tune large language models (LLMs). Its modular and plug-and-play nature allows the integration of various domain-specific LoRAs, enhancing LLM capabilities. Open-source platforms like Huggingface Modelscope have introduced a new computational paradigm, Uploadable Machine Learning (UML). In UML, contributors use decentralized data train specialized adapters, which are then uploaded central platform improve LLMs. This uses these...
In this article, a data-driven model based on the incremental deep extreme learning machine (IDELM) algorithm is proposed to predict temperature distribution in furnace. To end, computational fluid dynamics (CFD) simulations are carried out first get distributions under typical working conditions. Based air mode, simulation results divided into six subclasses. Then K-means clustering method applied find benchmark condition of each subclass. Moreover, random sampling used extract samples...
A novel method for the prediction of three-dimensional (3D) spatial distribution NOx in a furnace is proposed and evaluated. Computational fluid dynamics (CFD) simulations are conducted to generate data sets 3D distribution. The partitioned based on generation mechanisms improve model accuracy. Combining Pearson coefficient mutual information (PMI), input variables optimized by feature selection. established extreme learning machine (ELM). experiments considering 350 MW coal-fired boiler...
There is a fast-growing literature on estimating optimal treatment rules directly by maximizing the expected outcome. In biomedical studies and operations applications, censored survival outcome frequently observed, in which case restricted mean time probability are of great interest. this paper, we propose two robust criteria for learning with outcomes; former one targets at an rule time, where restriction specified given quantile such as median; latter buffered probabilities, predetermined...
Low-Rank Adaptation (LoRA) has emerged as a popular technique for fine-tuning large language models (LLMs) to various domains due its modular design and widespread availability on platforms like Huggingface. This modularity sparked interest in combining multiple LoRAs enhance LLM capabilities. However, existing methods LoRA composition primarily focus task-specific adaptations that require additional training, current model merging techniques often fail fully leverage LoRA's nature, leading...
Malaria epidemic along Thai-Myanmar border is still an ongoing occurrence. We explored malaria surveillance systems in Mae Sot District order to improve the detection and response efforts region. The main objective was study effectiveness of at border. Data were collected by reviewing medical records, interviewing personnel operation levels observing sites. reporting system under Bureau Epidemiology (BOE) hospital-based, with 76% coverage, 100% positive predictive value timeliness. It...
Sentiment analysis and emotion recognition are crucial tasks that utilize multimodal information. Transformer models have shown exceptional performance in fusion. However, traditional dot product transformers do not tolerate uncertainty inside sentiment data. In this study, we introduce rough set self-attention cross-attention mechanisms for recognition. A common concept is established based on granulation relations to extract important features through approximation. We then investigate a...
Numerous systems and applications have been established during the construction of Digital City, but there are two major obstacles preventing from being interactively collaboratively operated, one is data sharing, another service aggregation. The paper demonstrates a Common Platform city (CPGI) to provide common shared geographic information integrated services, promote sustainable collaborative operations all systems. Firstly CPGI defined, it compared with Geospatial Framework Fundamental...