Invention Application
- Patent Title: Neural Architecture Scaling For Hardware Accelerators
-
Application No.: US17175029Application Date: 2021-02-12
-
Publication No.: US20220230048A1Publication Date: 2022-07-21
- Inventor: Andrew Li , Sheng Li , Mingxing Tan , Ruoming Pang , Liqun Cheng , Quoc V. Le , Norman Paul Jouppi
- Applicant: Google LLC
- Applicant Address: US CA Mountain View
- Assignee: Google LLC
- Current Assignee: Google LLC
- Current Assignee Address: US CA Mountain View
- Main IPC: G06N3/04
- IPC: G06N3/04 ; G06N3/08

Abstract:
Methods, systems, and apparatus, including computer-readable media, for scaling neural network architectures on hardware accelerators. A method includes receiving training data and information specifying target computing resources, and performing using the training data, a neural architecture search over a search space to identify an architecture for a base neural network. A plurality of scaling parameter values for scaling the base neural network can be identified, which can include repeatedly selecting a plurality of candidate scaling parameter values, and determining a measure of performance for the base neural network scaled according to the plurality of candidate scaling parameter values, in accordance with a plurality of second objectives including a latency objective. An architecture for a scaled neural network can be determined using the architecture of the base neural network scaled according to the plurality of scaling parameter values.
Information query