Convolutional neural network optimization mechanism

US11727246B2 · kind B2 · utility

Assignee

• Intel Corporation · US

Inventors

• Liwei Ma · Beijing, CN
• Elmoustapha Ould-Ahmed-Vall · Chandler, US
• Barath Lakshmanan · Chandler, US
• Ben J. Ashbaugh · Folsom, US
• Jingyi Jin · Folsom, US
• Jeremy Bottleson · Rancho Cordova, US
• Mike B. Macpherson · Portland, US
• Kevin Nealis · San Jose, US
• Dhawal Srivastava · Phoenix, US
• Joydeep Ray · Folsom, US
• Ping T. Tang · Edison, US
• Michael S. Strickland · Sunnyvale, US
• Xiaoming Chen · Xi'an, CN
• Anbang Yao · Beijing, CN
• Tatiana Shpeisman · Menlo Park, US
• Altug Koker · Cameron Park, US
• Abhishek R. Appu · Cameron Park, US

Key dates

• Filing date: Feb 22, 2019
• Grant date: Aug 15, 2023
• Expiry date: Aug 13, 2039

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N3/045
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Embodiments provide systems and methods which facilitate optimization of a convolutional neural network (CNN). One embodiment provides for a non-transitory machine-readable medium storing instructions that cause one or more processors to perform operations comprising processing a trained convolutional neural network (CNN) to generate a processed CNN, the trained CNN having weights in a floating-point format. Processing the trained CNN includes quantizing the weights in the floating-point format to generate weights in an integer format. Quantizing the weights includes generating a quantization table to enable non-uniform quantization of the weights and quantizing the weights from the floating-point format to the integer format using the quantization table. The operations additionally comprise performing an inference operation utilizing the processed CNN with the integer format weights.