An energy-efficient convolution unit for depthwise separable convolutional neural networks

• Main Authors: Chong, Yi Sheng, Goh, Wang Ling, Ong, Yew-Soon, Nambiar, Vishnu P., Do, Anh Tuan
• Other Authors: Interdisciplinary Graduate School (IGS)
• Format: Conference or Workshop Item
• Language: English
• Published: 2021
• Subjects: Engineering::Electrical and electronic engineering; Convolutional Neural Network; CNN Accelerator
• Online Access: https://hdl.handle.net/10356/152096
• Institution: Nanyang Technological University

High performance but computationally expensive Convolutional Neural Networks (CNNs) require both algorithmic and custom hardware improvement to reduce model size and to improve energy efficiency for edge computing applications. Recent CNN architectures employ depthwise separable convolution to reduc...