公开(公告)号：US20210264279A1

公开(公告)日：2021-08-26

申请号：US16796397

申请日：2020-02-20

申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

发明人： Steve Esser ,  Jeffrey L. McKinstry ,  Deepika Bablani ,  Rathinakumar Appuswamy ,  Dharmendra S. Modha

IPC分类号： G06N3/08 ,  G06F17/16 ,  G06N3/04

摘要： Learned step size quantization in artificial neural network is provided. In various embodiments, a system comprises an artificial neural network and a computing node. The artificial neural network comprises: a quantizer having a configurable step size, the quantizer adapted to receive a plurality of input values and quantize the plurality of input values according to the configurable step size to produce a plurality of quantized input values, at least one matrix multiplier configured to receive the plurality of quantized input values from the quantizer and to apply a plurality of weights to the quantized input values to determine a plurality of output values having a first precision, and a multiplier configured to scale the output values to a second precision. The computing node is operatively coupled to the artificial neural network and is configured to: provide training input data to the artificial neural network, and optimize the configurable step size based on a gradient through the quantizer and the training input data.

公开(公告)号：US11823054B2

公开(公告)日：2023-11-21

申请号：US16796397

申请日：2020-02-20

申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

发明人： Steve Esser ,  Jeffrey L. McKinstry ,  Deepika Bablani ,  Rathinakumar Appuswamy ,  Dharmendra S. Modha

IPC分类号： G06F17/16 ,  G06F17/10 ,  G06N3/02 ,  G06N3/084 ,  G06N3/04

CPC分类号： G06N3/084 ,  G06F17/10 ,  G06F17/16 ,  G06N3/02 ,  G06N3/04

摘要： Learned step size quantization in artificial neural network is provided. In various embodiments, a system comprises an artificial neural network and a computing node. The artificial neural network comprises: a quantizer having a configurable step size, the quantizer adapted to receive a plurality of input values and quantize the plurality of input values according to the configurable step size to produce a plurality of quantized input values, at least one matrix multiplier configured to receive the plurality of quantized input values from the quantizer and to apply a plurality of weights to the quantized input values to determine a plurality of output values having a first precision, and a multiplier configured to scale the output values to a second precision. The computing node is operatively coupled to the artificial neural network and is configured to: provide training input data to the artificial neural network, and optimize the configurable step size based on a gradient through the quantizer and the training input data.