A5100432180- Bayesian Methods and Mixture Models
- Statistical Methods and Inference
- Bayesian Modeling and Causal Inference
- Soil Geostatistics and Mapping
- Hydrology and Drought Analysis
- Blind Source Separation Techniques
- Gaussian Processes and Bayesian Inference
- Financial Risk and Volatility Modeling
- Climate variability and models
- Anomaly Detection Techniques and Applications
- Adolescent Sexual and Reproductive Health
- Robotics and Sensor-Based Localization
- Protein Interaction Studies and Fluorescence Analysis
- HIV/AIDS Research and Interventions
- HIV/AIDS Impact and Responses
- Topic Modeling
- Sparse and Compressive Sensing Techniques
- Natural Language Processing Techniques
- Gene Regulatory Network Analysis
- Target Tracking and Data Fusion in Sensor Networks
- Neural Networks and Applications
- Coastal wetland ecosystem dynamics
- Remote Sensing and LiDAR Applications
- Spatial and Panel Data Analysis
- Deception detection and forensic psychology
Jiangnan University
2023-2025
State Key Laboratory of Food Science and Technology
2023-2025
Sun Yat-sen University
2023-2025
Dalian Maritime University
2025
Xidian University
2022-2025
Northwestern Polytechnical University
2022-2024
Shanghai Jiao Tong University
2022-2024
Henan University of Technology
2023-2024
Shanghai University of Engineering Science
2024
Chengdu Women's and Children's Central Hospital
2024
Crustal rock strength from outer space The response of crustal to stresses is challenging estimate yet vital for determining risks events such as earthquakes. Moore et al. take advantage the recent M w 7.1 Kumamoto earthquake in Japan determine rheology rocks region. observed inversion strain rates demonstrates that certain areas have stiff and others (e.g., under Aso volcanic complex) much weaker rock. results match up with expectations, which means method can successfully measure...
This study was aimed to evaluate the performance of DESs functionalized peanut shell (PSD) as biosorbent for removing Cr(VI) from water. The effects pretreatment, initial concentration, adsorption temperature, kinetics, isotherm, and thermodynamics were investigated. Scanning electron microscopy (SEM) Point Zero charge (pHpzc) techniques used characterization adsorbents. results showed that rigid structure broken down after modification point zero 6.02 6.84 PSD, which exhibited a slightly...
Smooth handling of pedestrian interactions is a key requirement for Autonomous Vehicles (AV) and Advanced Driver Assistance Systems (ADAS). Such systems call early accurate prediction pedestrian's crossing/not-crossing behaviour in front the vehicle. Existing approaches to make use motion, his/her location scene static context variables such as traffic lights, zebra crossings etc. We stress on necessity smooth operation systems. introduce influence vehicle intention this purpose. In paper,...
In recent decades, emotion recognition has received considerable attention. As more enthusiasm shifted to the physiological pattern, a wide range of elaborate data features come up and are combined with various classifying models detect one's emotional states. To circumvent labor artificially designing features, we propose acquire affective robust representations automatically through Stacked Denoising Autoencoder (SDA) architecture unsupervised pre-training, followed by supervised...
Code Large Language Models (Code LLMs) have gained significant attention in the industry due to their wide applications full lifecycle of software engineering. However, effectiveness existing models understanding non-English inputs for multi-lingual code-related tasks is still far from well studied. This paper introduces CodeFuse-13B, an open-sourced pre-trained code LLM 2. It specifically designed with both English and Chinese prompts supports over 40 programming languages. CodeFuse...
Graphene has garnered significant attention due to its unique and remarkable properties. The widespread application of graphene materials in numerous fields inevitably leads their release into the environment. This study examines long-term impacts on anaerobic sequencing batch reactors. low-concentration (5 mg L-1) exhibited a inhibitory effect removal chemical oxygen demand, while high-concentration group (100 was less affected. transmission electron microscopy Raman spectroscopy results...
Ship detection technology represents a significant research focus within the application domain of synthetic aperture radar. Among all methods, deep learning method stands out for its high accuracy and efficiency. However, large-scale algorithm training requires huge computing power support large equipment to process, which is not suitable real-time on edge platforms. Therefore, achieve fast data transmission little computation complexity, design lightweight models becomes hot point. In...
Water body segmentation is an important tool for the hydrological monitoring of Earth. With rapid development convolutional neural networks, semantic techniques have been used on remote sensing images to extract water bodies. However, some difficulties need be overcome achieve good results in segmentation, such as complex background, huge scale, connectivity, and rough edges. In this study, a model (DUPnet) with dense connectivity multi-scale pyramidal pools proposed rapidly accurately...
Gaussian hidden variable graphical models are powerful tools to describe high-dimensional data; they capture dependencies between observed (Gaussian) variables by introducing a suitable number of variables. However, such only applicable data. Moreover, sensitive the choice certain regularization parameters. In this paper, (1) copula introduced, which extend non-Gaussian (2) sparsity pattern model is learned via stability selection, leads more stable results than cross-validation and other...
Assessing the risk of extreme events in a spatial domain, such as hurricanes, floods, and droughts, presents unique significance practice. Unfortunately, existing extreme-value statistical models are typically not feasible for practical large-scale problems. Graphical models, on other hand, capable handling sizable number variables, but have yet to be explored realm analysis. To bridge gap, an graphical model is introduced this paper, i.e., ensemble-of-trees pairwise copulas (ETPC). In...
Dynamic compressed sensing (DCS) has recently gained popularity as a successful approach to recovering dynamic sparse signals. In this paper, we attack the problem from Bayesian perspective. The proposed model imposes constraints on both unknown signal and its temporal innovation via t priors. Due conjugacy between priors likelihoods, are able propose computationally efficient mean-field variational Bayes algorithm learn without parameter tuning. We consider online offline scenarios,...
Human lying is influenced by cognitive neural mechanisms in the brain, and conducting research on lie detection speech can help to reveal of human brain. Inappropriate deception features easily lead dimension disaster make generalization ability widely used semi-supervised model worse. Because this, this paper proposes a algorithm combining acoustic statistical time-frequency two-dimensional features. Firstly, hybrid network based autoencoder (AE) mean-teacher established. Secondly, static...
Copula Gaussian graphical models are capable of describing dependencies between a large number heterogeneous variables. In this paper, low-complexity algorithms proposed for learning copula from discrete data. The approach is Monte-Carlo expectation maximization: in the E-step, an efficient Gibbs sampler applied, and M-step, sparse model inferred by solving penalized maximize likelihood problem. regularization parameter determined through BINCO method Li et al. Numerical results both...
To assess the risk of extreme events such as hurricanes, earthquakes, and floods, it is crucial to develop accurate extreme-value statistical models. Extreme often display heterogeneity (i.e., nonstationarity), varying continuously with a number covariates. Previous studies have suggested that models considering covariate effects lead reliable estimates distributions. In this paper, we novel model incorporate multiple Specifically, analyze an example sea states in Gulf Mexico, where...