A neural network-based on-device learning anomaly detector for edge devices

Mathematics

• Concept Drift 100%
• Detector 61%
• Anomaly 59%
• Power Consumption 55%
• Server 55%
• Data Transfer 55%
• Anomaly Detection 53%
• Neural Networks 48%
• Learning 47%
• Latency 45%
• System-on-chip 33%
• Back-propagation Neural Network 30%
• Field Programmable Gate Array 27%
• Energy Consumption 26%
• Training 18%
• Software 15%
• Optimization 13%
• Experiment 12%
• Evaluation 12%
• Generalization 8%
• Model 5%

Engineering & Materials Science

• Detectors 40%
• Servers 37%
• Anomaly detection 35%
• Data transfer 35%
• Neural networks 31%
• Electric power utilization 25%
• Intellectual property core 21%
• System-on-chip 17%
• Backpropagation 16%
• Program processors 14%
• Field programmable gate arrays (FPGA) 14%
• Energy utilization 12%
• Experiments 6%