Automated machine learning pipeline generation

US11620582B2 · kind B2 · utility

Assignee

• International Business Machines Corporation · US

Inventors

• Bei Chen · Dublin, IE
• Long H. Vu · White Plains, US
• Syed Yousaf Shah · Yorktown Heights, US
• Xuan-Hong Dang · Chappaqua, US
• Peter D. Kirchner · Garrison, US
• Si Er Han · Xi'an, CN
• Ji Hui Yang · Beijing, CN
• Jun Wang · Nanjing, CN
• Jing Xu · Shanghai, CN
• Dakuo Wang · Cambridge, US
• Dhavalkumar C. Patel · White Plains, US
• Gregory Bramble · Larchmont, US
• Horst Cornelius Samulowitz · Armonk, US
• Saket Sathe · Mohegan Lake, US
• Chuang Gan · Cambridge, US

Key dates

• Filing date: Jul 29, 2020
• Grant date: Apr 4, 2023
• Expiry date: May 30, 2041

Classification

• Technology area (CPC G): Physics
• CPC primary: G06N20/10
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Techniques regarding one or more automated machine learning processes that analyze time series data are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a time series analysis component that selects a machine learning pipeline for meta transfer learning on time series data by sequentially allocating subsets of training data from the time series data amongst a plurality of machine learning pipeline candidates.