PAC-Bayesian framework based drop-path method for 2D discriminative convolutional network pruning

• Published in: Multidimensional systems and signal processing Vol. 31; no. 3; pp. 793 - 827
• Main Authors: Zheng, Qinghe, Tian, Xinyu, Yang, Mingqiang, Wu, Yulin, Su, Huake
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2020; Springer Nature B.V
• Subjects: Artificial Intelligence; Artificial neural networks; Bayesian analysis; Circuits and Systems; Electrical Engineering; Engineering; Image classification; Mathematical models; Model accuracy; Multilayers; Object recognition; Parameters; Pruning; Signal,Image and Speech Processing; Two dimensional models; Visual tasks; Deep learning; Model acceleration; Structure exploration; Model compression; Generalization error boundary

Deep convolutional neural networks (CNNs) have demonstrated its extraordinary power on various visual tasks like object detection and classification. However, it is still challenging to deploy state-of-the-art models into real-world applications, such as autonomous vehicles, due to their expensive c...