公开(公告)号：US20240054332A1

公开(公告)日：2024-02-15

申请号：US18496411

申请日：2023-10-27

Applicant: Advanced Micro Devices, Inc.

Inventor： Daniel I. Lowell ,  Sergey Voronov ,  Mayank Daga

IPC: G06N3/063 ,  G06N3/08

CPC classification number: G06N3/063 ,  G06N3/08

Abstract: Methods, devices, systems, and instructions for adaptive quantization in an artificial neural network (ANN) calculate a distribution of ANN information; select a quantization function from a set of quantization functions based on the distribution; apply the quantization function to the ANN information to generate quantized ANN information; load the quantized ANN information into the ANN; and generate an output based on the quantized ANN information. Some examples recalculate the distribution of ANN information and reselect the quantization function from the set of quantization functions based on the resampled distribution if the output does not sufficiently correlate with a known correct output. In some examples, the ANN information includes a set of training data. In some examples, the ANN information includes a plurality of link weights.

公开(公告)号：US11803734B2

公开(公告)日：2023-10-31

申请号：US15849617

申请日：2017-12-20

Applicant: Advanced Micro Devices, Inc.

Inventor： Daniel I. Lowell ,  Sergey Voronov ,  Mayank Daga

IPC: G06N3/063 ,  G06N3/08

CPC classification number: G06N3/063 ,  G06N3/08

Abstract: Methods, devices, systems, and instructions for adaptive quantization in an artificial neural network (ANN) calculate a distribution of ANN information; select a quantization function from a set of quantization functions based on the distribution; apply the quantization function to the ANN information to generate quantized ANN information; load the quantized ANN information into the ANN; and generate an output based on the quantized ANN information. Some examples recalculate the distribution of ANN information and reselect the quantization function from the set of quantization functions based on the resampled distribution if the output does not sufficiently correlate with a known correct output. In some examples, the ANN information includes a set of training data. In some examples, the ANN information includes a plurality of link weights.

公开(公告)号：US20190188557A1

公开(公告)日：2019-06-20

申请号：US15849617

申请日：2017-12-20

Applicant: Advanced Micro Devices, Inc.

Inventor： Daniel I. Lowell ,  Sergey Voronov ,  Mayank Daga

IPC: G06N3/063 ,  G06N3/08

CPC classification number: G06N3/063 ,  G06N3/08

Abstract: Methods, devices, systems, and instructions for adaptive quantization in an artificial neural network (ANN) calculate a distribution of ANN information; select a quantization function from a set of quantization functions based on the distribution; apply the quantization function to the ANN information to generate quantized ANN information; load the quantized ANN information into the ANN; and generate an output based on the quantized ANN information. Some examples recalculate the distribution of ANN information and reselect the quantization function from the set of quantization functions based on the resampled distribution if the output does not sufficiently correlate with a known correct output. In some examples, the ANN information includes a set of training data. In some examples, the ANN information includes a plurality of link weights.