A5026950991- Adversarial Robustness in Machine Learning
- Wastewater Treatment and Nitrogen Removal
- Software Testing and Debugging Techniques
- Metallurgical Processes and Thermodynamics
- High-Temperature Coating Behaviors
- Railway Systems and Energy Efficiency
- Advanced Multi-Objective Optimization Algorithms
- Anomaly Detection Techniques and Applications
- Transportation Planning and Optimization
- Robotics and Sensor-Based Localization
- Evolutionary Algorithms and Applications
- Advanced Neural Network Applications
- Fiber-reinforced polymer composites
- Advanced Steganography and Watermarking Techniques
- Railway Engineering and Dynamics
- Chaos-based Image/Signal Encryption
- Water Treatment and Disinfection
- Scheduling and Optimization Algorithms
- Metallurgy and Material Forming
- Microstructure and Mechanical Properties of Steels
- Catalytic Processes in Materials Science
- Metaheuristic Optimization Algorithms Research
- Economic and Technological Innovation
- Plant Disease Resistance and Genetics
- Microbial Fuel Cells and Bioremediation
Beijing Jiaotong University
2019-2025
Nanjing University
2019-2024
Zhejiang University
2024
Institute of Soil Science
2024
University of Nottingham
2024
Harbin Medical University
2024
Second Affiliated Hospital of Harbin Medical University
2024
University of Delaware
2023
Xi'an University of Technology
2023
China National Space Administration
2023
Deep neural networks (DNN) have been deployed in many software systems to assist various classification tasks. In company with the fantastic effectiveness classification, DNNs could also exhibit incorrect behaviors and result accidents losses. Therefore, testing techniques that can detect DNN improve quality are extremely necessary critical. However, oracle, which defines correct output for a given input, is often not available automated testing. To obtain oracle information, tasks of...
Deep neural networks (DNN) have achieved tremendous development in the past decade. While many DNN-driven software applications been deployed to solve various tasks, they could also produce incorrect behaviors and result massive losses. To reveal improve quality of applications, developers often need rich labeled data for testing optimization DNN models. However, practice, collecting diverse from application scenarios labeling them properly is a highly expensive time-consuming task.
This paper aims to develop a dynamic, real-time scheduling strategy under interference that can minimize the negative impact of on production without sacrificing efficiency.Taking minimal cost and makespan as objectives optimization function, author put forward parallel hybrid algorithm for rescheduling interference, aiming strike balance between processing disturbance.The benchmark test results show proposed achieved better accuracy than NSGA-II AMOSA, its has nothing do with distribution...
Most tabia sites exhibit accelerated decay rates caused by weathering. Weathering damage generally proceeds from the surface to interior; thus, protecting of is an effective method resist In this study, microbially induced carbonate precipitation (MICP) technique used form anti-erosion layer on tabia. Next, applicability MICP for erosion control examined in laboratory via static contact angle test, capillary water absorption Karsten tube durability acid resistance vapor permeability (WVP)...
Multi-Sensor Fusion (MSF) based perception systems have been the foundation in supporting many industrial applications and domains, such as self-driving cars, robotic arms, unmanned aerial vehicles. Over past few years, fast progress datadriven artificial intelligence (AI) has brought a fast-increasing trend to empower MSF by deep learning techniques further improve performance, especially on intelligent their systems. Although quite AI-enabled proposed, up present, limited benchmarks that...
Multi-sensor fusion stands as a pivotal technique in addressing numerous safety-critical tasks and applications, e.g., self-driving cars automated robotic arms. With the continuous advancement data-driven artificial intelligence (AI), MSF's potential for sensing understanding intricate external environments has been further amplified, bringing profound impact on intelligent systems specifically their perception systems. Similar to traditional software, adequate testing is also required...
Distant metastasis is a prevalent cause of mortality in gastric cancer (GC) patients. Anoikis, process that induces cell death when cells get detached from the extracellular matrix (ECM), acts as barrier to tumor metastasis. To survive circulatory system and metastasize, must acquire anoikis resistance. It crucial identify molecular processes resistance GC since this might lead discovery novel treatment targets improve long-term survival In study, we employed quantitative proteomics growth...
Based on the measurement of producer service industry agglomeration and export technological complexity manufactured products in 288 Chinese cities from 2000 to 2015, this paper illustrates evolvement spatial characteristics two factors through visualization figures, discusses effects services manufacturing robust panel data models. The findings are as follows: with influence industrial connection, empirical outcomes indicate that urban can promote full-sample level. Visualization analysis...
Herein, the effect of different contents Cr on isothermal structure transformation FeO is systematically studied by a thermogravimetric analyzer, and it found that temperature range eutectoid in Fe–Cr alloy 300–500 °C. Combined with experimental data, kinetic model phase established based Johnson–Mehl–Avrami–Kolmogorov equation, time–temperature–transformation curve predicted, which follows law C curve, shifts backward increase content. Therm‐calc thermodynamic software calculates Fe–O...
This article builds a stress–strain prediction model based on production data from the steel industry by using machine learning algorithms. Based of 9Ni hot deformation behavior, flow stress constitutive equation is established. Four models, including Arrhenius-type considering strain compensation, Stochastic Configuration Networks (SCNs) neural network, Multi-objective Particle Swarm Optimization (AMPSO) and Support Vector Machine (SVM) model, are adopted in this research. The results show...
Abstract Multifactorial optimization (MFO) is a kind of problem that has attracted considerable attention in recent years. The multifactorial evolutionary algorithm utilizes the implicit genetic transfer mechanism characterized by knowledge to conduct multitasking simultaneously. Therefore, effectiveness significantly affects performance algorithm. To achieve positive transfer, this paper proposed an with adaptive strategy based on decision tree (EMT-ADT). evaluate useful contained...