公开(公告)号：US12131258B2

公开(公告)日：2024-10-29

申请号：US17030315

申请日：2020-09-23

Applicant: QUALCOMM Incorporated

Inventor： Yadong Lu ,  Ying Wang ,  Tijmen Pieter Frederik Blankevoort ,  Christos Louizos ,  Matthias Reisser ,  Jilei Hou

IPC: G06N3/082 ,  G06N3/04 ,  G06N3/063

CPC classification number: G06N3/082 ,  G06N3/04 ,  G06N3/063

Abstract: A method for compressing a deep neural network includes determining a pruning ratio for a channel and a mixed-precision quantization bit-width based on an operational budget of a device implementing the deep neural network. The method further includes quantizing a weight parameter of the deep neural network and/or an activation parameter of the deep neural network based on the quantization bit-width. The method also includes pruning the channel of the deep neural network based on the pruning ratio.

公开(公告)号：US11790241B2

公开(公告)日：2023-10-17

申请号：US17016130

申请日：2020-09-09

Applicant: QUALCOMM Incorporated

Inventor： Matthias Reisser ,  Saurabh Kedar Pitre ,  Xiaochun Zhu ,  Edward Harrison Teague ,  Zhongze Wang ,  Max Welling

IPC: G06F17/16 ,  G06N3/10 ,  G06N3/047

CPC classification number: G06N3/10 ,  G06F17/16 ,  G06N3/047

Abstract: In one embodiment, a method of simulating an operation of an artificial neural network on a binary neural network processor includes receiving a binary input vector for a layer including a probabilistic binary weight matrix and performing vector-matrix multiplication of the input vector with the probabilistic binary weight matrix, wherein the multiplication results are modified by simulated binary-neural-processing hardware noise, to generate a binary output vector, where the simulation is performed in the forward pass of a training algorithm for a neural network model for the binary-neural-processing hardware.

公开(公告)号：US11562208B2

公开(公告)日：2023-01-24

申请号：US16413535

申请日：2019-05-15

Applicant: QUALCOMM Incorporated

Inventor： Christos Louizos ,  Matthias Reisser ,  Tijmen Pieter Frederik Blankevoort ,  Max Welling

IPC: G06N3/04 ,  G06N3/08 ,  G06F17/18

Abstract: A method for quantizing a neural network includes modeling noise of parameters of the neural network. The method also includes assigning grid values to each realization of the parameters according to a concrete distribution that depends on a local fixed-point quantization grid and the modeled noise and. The method further includes computing a fixed-point value representing parameters of a hard fixed-point quantized neural network.