Abstract
Computational complexity of Convolutional Neural Networks (CNNs) makes its integration in embedded systems a challenging task. Methods allowing to simplify these algorithms are therefore of great interest. In this paper, we propose a new CNN compression method based on the application of Principal Component Analysis (PCA) in a layer-wise fashion, and show the benefits of an additional fine-tuning step. Through this method, it is possible to reach very flexible trade-offs between network size and accuracy, such as a x2 reduction in the number of parameters for an accuracy drop inferior to 2%. We also discuss its compatibility with other well-known CNN compression methods.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
