公开(公告)号：US20220351033A1

公开(公告)日：2022-11-03

申请号：US17242987

申请日：2021-04-28

Applicant: Arm Limited

Inventor： Paul Nicholas WHATMOUGH ,  Mark John O'CONNOR

IPC: G06N3/08 ,  G06N3/04 ,  G06K9/62

Abstract: A method of operating a system having a plurality of neural networks includes receiving sequential input data events and processing each sequential input data event using a corresponding subset of the plurality of neural networks to obtain a plurality of sequential outputs. Each sequential output is indicative of a predictive determination of an aspect of the corresponding input data event. The method includes processing the plurality of sequential outputs to determine an uncertainty value associated with the plurality of sequential outputs, and operating the system based on the determined uncertainty value.

公开(公告)号：US20230073669A1

公开(公告)日：2023-03-09

申请号：US18055192

申请日：2022-11-14

Applicant: Arm Limited

Inventor： Mark John O'CONNOR

IPC: G06N3/08 ,  G06N3/04

Abstract: A computer-implemented method of optimising a student neural network (SNN), based on a previously-trained neural network (PTNN) trained on first data (FD) using a first processing system (FPS). The method includes using a second processing system (SPS) to generate reference output data (ROD) from the previously-trained neural network (PTNN) in response to inputting second data (SD) to the previously-trained neural network (PTNN). The method also includes optimising a student neural network (SNN) for processing the second data (SD) with the second processing system (SPS), by using the second processing system (SPS) to adjust a plurality of parameters of the student neural network (SNN) such that a difference (DIFF) between the reference output data (ROD), and second output data (SOD) generated by the student neural network (SNN) in response to inputting the second data (SD) to the student neural network (SNN), satisfies a stopping criterion.

公开(公告)号：US20240028877A1

公开(公告)日：2024-01-25

申请号：US17870038

申请日：2022-07-21

Applicant: Arm Limited

Inventor： Shounak DATTA ,  Dibakar GOPE ,  Jesse Garrett BEU ,  Mark John O'CONNOR

IPC: G06N3/063

CPC classification number: G06N3/063

Abstract: There is provided a neural processing unit for calculating an attention matrix during machine learning inference. The neural processing unit is configured to calculate: a first score matrix based on differences between a query matrix and a key matrix; a second score matrix based on differences between the key matrix and a learned key matrix; a similarity matrix based on a combination of the first score matrix and second score matrix; and an attention matrix comprising applying a normalisation function to the similarity matrix. Also provided is an apparatus comprising at least one said neural processing unit and at least one memory, the memory configured to pass, on demand, a learned key matrix to the neural processing unit. Also provided is a computer program product having computer readable program code stored thereon which, when executed by said neural processing unit, causes the unit to perform said calculations.

公开(公告)号：US20220121927A1

公开(公告)日：2022-04-21

申请号：US17076392

申请日：2020-10-21

Applicant: Arm Limited

Inventor： Mark John O'CONNOR

IPC: G06N3/08 ,  G06N3/04 ,  G06N3/063 ,  G06K9/62 ,  G06F11/34

Abstract: A computer-implemented method of providing a group of neural networks for processing data includes: identifying a group of neural networks including a main neural network and one or more sub-neural networks, each neural network comprising a plurality of parameters and wherein one or more of the parameters of each sub-neural network are shared by the sub-neural network and the main neural network; inputting training data into each neural network, and adjusting the parameters of each neural network; computing a performance score for each neural network using the adjusted parameters; generating a combined score for the group of neural networks by combining the performance score, with a value of a loss function computed for each neural network using the adjusted parameters; repeating the identifying and the inputting and the adjusting and the computing and the generating; and selecting a group of neural networks for processing data in the plurality of hardware environments based on the value of the combined score for each group of neural networks.