A5100717593- Advanced Numerical Analysis Techniques
- Advanced Measurement and Metrology Techniques
- Advanced machining processes and optimization
- Advanced Malware Detection Techniques
- Advanced Surface Polishing Techniques
- Advanced Image Processing Techniques
- Advanced Image and Video Retrieval Techniques
- Robotics and Sensor-Based Localization
- Image Processing Techniques and Applications
- Software Testing and Debugging Techniques
- Advanced Neural Network Applications
- Machine Learning and ELM
- Adversarial Robustness in Machine Learning
- Software Engineering Research
- Advanced Vision and Imaging
- Artificial Intelligence in Games
- Thermoregulation and physiological responses
- Anomaly Detection Techniques and Applications
- Gyrotron and Vacuum Electronics Research
- RFID technology advancements
- Biochemical and Structural Characterization
- Hate Speech and Cyberbullying Detection
- E-commerce and Technology Innovations
- Domain Adaptation and Few-Shot Learning
- Radiomics and Machine Learning in Medical Imaging
Huazhong University of Science and Technology
2019-2024
City College of Dongguan University of Technology
2021-2024
Nanjing University of Chinese Medicine
2023
Xi'an University of Technology
2021
University of Science and Technology Beijing
2020
Hubei University
2020
Minzu University of China
2019
South China University of Technology
2018
State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering
2010
Abstract With the globalization of global economy, protection intangible cultural heritage is facing a crisis survival. Intangible crystallization wisdom working people country in past dynasties, with strong national characteristics. Applying digital technology to can promote inheritance, transformation and application heritage. Since China put forward information development strategy, it has become more practical use protect Due imperfection management mechanism format specification, as...
Recent methods have demonstrated that machine learning (ML) based static malware detection models are vulnerable to adversarial attacks. However, the generated often fails generalize production-level anti-malware software (AMS), as they usually involve multiple methods. This calls for universal solutions problem of variants generation. In this work, we demonstrate how proposed method, MalwareTotal, has allowed continue abound in ML-based, signature-based, and hybrid software. Given a...
In the RFID system, tag collision problem seriously affects efficiency of identification. Among anti-collision algorithms, tree-based algorithms are popular for reason that they can ensure tags theoretically 100% be identified by readers. As multi-reader scenarios more and widely used in complex internet things environments, there slots idle traditional which algorithm. We propose a interactive algorithm (TMIA) identification to solve problems above. Readers optimize broadcast prefix...
The vast majority of indoor navigation algorithms either rely on manual scene augmentation and labelling or exploit multi-sensor fusion techniques in achieving simultaneous localization mapping (SLAM), leading to high computational costs, hardware complexities robustness deficiencies. This paper proposes an efficient robust deep learning-based framework for robots. Firstly, we put forward end-to-end trainable siamese convolutional neural network (DCNN) which decomposes into orientation one...
Researches on computer games for Go, Chess, and Japanese Chess stand out as one of the notable landmarks in progress artificial intelligence. AlphaGo, AlphaGo Zero, AlphaZero algorithms, which are called style (AZ-style) algorithms some literature [1], have achieved superhuman performance by using deep reinforcement learning (DRL). However, unavailability training details, expensive equipment used model training, low evaluation accuracy resulted slow self-play without computing practical...
Abstract Using multi‐task learning to extract code features can effectively increase the information of features. However, existing methods mainly have two limitations: (1) They cannot enough code‐related or only similar semantic Similar makes in increased insufficiently. high difference is challenging converge. (2) train on heterogeneous datasets. In standard training, we need label all tasks for data, which consumes enormous labor. To solve above limitations, select tasks, cross‐language...
Abstract The continuous development of the information age promotes transformation teaching methods, and forms students’ learning receiving knowledge are also constantly updated. In this paper, virtual reality technology is used to realize deep integration traditional culture ideological political education from three aspects course: objectives, content process, design a three-dimensional animation classroom. particle swarm algorithm optimize acquisition resources integrate best that...
Abstract It has gradually become a trend to infiltrate the knowledge of non-heritage culture and inherit spirit ethnic crafts in art education colleges universities. The purpose this paper is discuss significance universities, problems faced by integration into transmission path We hope enhance deep connection between college increasing research efforts, so that students will pay attention preservation pass on studies. This develops cultivation intangible cultural heritage through...
Target-guided navigation establishes the foundation for efficiently addressing vision-based multi-agent coordination robotics. This work proposes a multi-objective deep convolution network which consists of two parallel branches built atop shared feature extractor. The proposed is capable concurrently constructing semantic maps while achieving efficient visual detection designated guider robot or landmark towards outdoor navigation. In order to achieve low latency requirements controller,...