A5027803367- Advanced Neural Network Applications
- Anomaly Detection Techniques and Applications
- Adversarial Robustness in Machine Learning
- Brain Tumor Detection and Classification
- Machine Learning and ELM
- IoT and Edge/Fog Computing
- Vehicle License Plate Recognition
- Machine Learning and Data Classification
- Autonomous Vehicle Technology and Safety
- Advanced Memory and Neural Computing
- Network Security and Intrusion Detection
- Advanced Malware Detection Techniques
- Vehicular Ad Hoc Networks (VANETs)
University of Science and Technology of China
2023
Suzhou University of Science and Technology
2020-2022
Suzhou Research Institute
2020-2021
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ResNets</i> are widely used in the intrusion detection system (IDS) of software-defined industrial network to construct accurate intelligence attacks. However, IDS based on ResNets has a long detecting interval because fine-grained operator and intermediate outcomes multi-branch architecture ResNets. To address this problem, article, we propose Juggler-ResNet with fusible residual structure that...
Abstract Because of their portability, electric motorcycles are usually pushed into elevators by residents and charged in the home, which has serious safety risks. Traditional manual-based methods to manage this behavior have poor monitoring effects high costs. As for automatic management systems using artificial intelligence (AI), deployment method matters. Cloud-based disadvantages latency, risk privacy leakage, heavy network transmission loads. In paper, we propose a highly secure...
Object detection at the edge side is a common task in various environments. The deployment of convolutional neural networks intelligent systems very challenging because highly constrained main-memory space. This study aims operating with reduced memory requirement. basic idea that tasks same type would involve critical subnetwork. We propose identifying network connections by considering importance channels. During runtime, proposed method detects types and timely swaps model parameters...
Deep Neural Network (DNN) models have brought significant performance gains to machine learning tasks. Nevertheless, the huge storage and computational costs of high-performance severely limit their deployment on resource-limited embedded devices. Knowledge distillation (KD) is mainstream approach for compression, but existing KD framework has some limitations. First, student model underperforms when there a considerable capacity gap between teacher models, which reduces compression rate in...