Invention Grant
- Patent Title: Connection weight learning for guided architecture evolution
-
Application No.: US17605783Application Date: 2020-05-22
-
Publication No.: US12046025B2Publication Date: 2024-07-23
- Inventor: Michael Sahngwon Ryoo , Anthony Jacob Piergiovanni , Mingxing Tan , Anelia Angelova
- Applicant: Google LLC
- Applicant Address: US CA Mountain View
- Assignee: Google LLC
- Current Assignee: Google LLC
- Current Assignee Address: US CA Mountain View
- Agency: Fish & Richardson P.C.
- International Application: PCT/US2020/034267 2020.05.22
- International Announcement: WO2020/237168A 2020.11.26
- Date entered country: 2021-10-22
- Main IPC: G06V10/82
- IPC: G06V10/82 ; G06N3/045 ; G06T1/20 ; G06T3/4046 ; G06T7/207 ; G06V10/776

Abstract:
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining one or more neural network architectures of a neural network for performing a video processing neural network task. In one aspect, a method comprises: at each of a plurality of iterations: selecting a parent neural network architecture from a set of neural network architectures; training a neural network having the parent neural network architecture to perform the video processing neural network task, comprising determining trained values of connection weight parameters of the parent neural network architecture; generating a new neural network architecture based at least in part on the trained values of the connection weight parameters of the parent neural network architecture; and adding the new neural network architecture to the set of neural network architectures.
Public/Granted literature
- US20220189154A1 CONNECTION WEIGHT LEARNING FOR GUIDED ARCHITECTURE EVOLUTION Public/Granted day:2022-06-16
Information query