Approximate Acceleration for CNN-based Applications on IoT Edge Devices

Engineering & Materials Science

• Convolutional neural networks 100%
• Energy utilization 19%
• Internet of things 92%
• Machine learning 20%
• Particle accelerators 55%
• Reconfigurable hardware 36%
• System-on-chip 27%

Physics & Astronomy

• accelerators 44%
• electric batteries 28%
• energy consumption 32%
• machine learning 34%
• output 16%
• reconfigurable hardware 53%
• resilience 37%
• resources 54%
• systems-on-a-chip 40%