Automated generation of machine learning models

US11348032B1 · kind B1 · utility

Assignee

• Meta Platforms, Inc. · US

Inventors

• Jurgen Anne Francois Marie Van Gael · Cambridge, GB
• Yu Ning · Sunnyvale, US
• Hao Shi · 安丰镇, CN
• Fei Xie · Sunnyvale, US
• Bingyue Peng · San Jose, US
• Shyamsundar Rajaram · San Francisco, US
• Xin Liu · Belmont, US
• Zhen Yao · Shanghai, CN
• Peng Yang · Chesterfield, US
• Robert Oliver Burns Zeldin · Los Altos, US
• Piyush Bansal · Menlo Park, US

Key dates

• Filing date: Sep 2, 2018
• Grant date: May 31, 2022
• Expiry date: Nov 23, 2040

Classification

• Technology area (CPC G): Physics
• CPC primary: G06F8/31
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Machine-trained models are generated based on a model description that defines parameters for training the model and that can inherit parameters from parent model descriptions. When a parent model description changes, the changes made to the parent model description are applied to the model description automatically. When a target model is re-generated, a description of the set of parameters for generating the target model is received. The parent model is then identified from the received description, and a description of the set of parameters for generating the parent model is retrieved. Using the description for the target model and the parent model, a pipeline for generating the target model is generated. Finally, the pipeline is executed to generate the target model.