公开(公告)号：WO2023075742A1

公开(公告)日：2023-05-04

申请号：PCT/US2021/056418

申请日：2021-10-25

Applicant: GOOGLE LLC

Inventor： KUMAR, Tushar ,  HALAMBI, Soorgoli Ashok ,  PARK, Jason Jong Kyu ,  CHAUHAN, Arun ,  WOO, Dong Hyuk

IPC: G06N3/04 ,  G06N3/063

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing inference computations of a fully convolutional neural network receiving inputs with different sizes. One of the methods include receiving a new input to be processed by a fully convolutional neural network, the new input having a first size different from a fixed size that the fully convolutional neural network is configured to process; determining, one or more fixed-size inputs from the new input, each fixed-size input having the fixed size; obtaining a respective fixed-size output generated by the fully convolutional neural network performing inference computations for each of the one or more fixed-size inputs; and generating, from the respective fixed-size outputs comprising one or more invalid pixel values, a final output that is equivalent to an output that would be generated by processing the new input using the fully convolutional neural network.

公开(公告)号：WO2022139795A1

公开(公告)日：2022-06-30

申请号：PCT/US2020/066430

申请日：2020-12-21

Applicant: GOOGLE LLC

Inventor： FADELU, Temitayo ,  NARAYANASWAMI, Ravi ,  MIN, JiHong ,  LI, Dongdong ,  GUPTA, Suyog ,  PARK, Jason Jong Kyu

IPC: G06F9/48

Abstract: The present disclosure describes a system and method for preempting a long-running process with a higher priority process in a machine learning system, such as a hardware accelerator. The machine learning hardware accelerator can be a multi-chip system including semiconductor chips that can be application-specific integrated circuits (ASIC) designed to perform machine learning operations. An ASIC is an integrated circuit (IC) that is customized for a particular use.

公开(公告)号：WO2021126194A1

公开(公告)日：2021-06-24

申请号：PCT/US2019/067289

申请日：2019-12-18

Applicant: GOOGLE LLC

Inventor： LIU, Jack ,  WOO, Dong Hyuk ,  PARK, Jason Jong Kyu ,  ASHOK, Raksit

IPC: G06N3/063 ,  G06N3/04 ,  G06N3/0454

Abstract: Methods and systems, including computer programs encoded on a computer storage medium. In one aspect, a method includes obtaining data specifying one or more neural networks to be deployed on a neural network hardware accelerator, each of the one or more neural networks having a respective set of parameters, and the neural network hardware accelerator having one or more memories having a memory capacity; determining a maximum amount of the memory capacity that will be in use at any one time during a processing of any of the one or more neural networks by the neural network hardware accelerator; identifying a subset of the parameters of the one or more neural networks that consumes an amount of memory that is less than a difference between the memory capacity and the determined maximum amount of the memory capacity; and storing the identified subset of the parameters.