公开(公告)号：US20250139193A1

公开(公告)日：2025-05-01

申请号：US18820372

申请日：2024-08-30

Applicant: Samsung Electronics Co., Ltd.

Inventor： Younho JEON ,  Jae Hun JANG ,  Suchang KIM ,  Yeo-Reum PARK ,  Hong Rak SON ,  Jihoon LIM ,  Se Jung KWON ,  Byeoung Wook KIM ,  Baeseong PARK ,  Dongsoo LEE

IPC: G06F17/16

Abstract: A matrix multiplier includes an input vector scaler generating a first quantization scaled input vector based on a first input vector, a plurality of common scale coefficients, and first-to-Rth multiplication scale coefficients, a first data type converter generating a first fixed point quantization scaled input vector based on the first quantization scaled input vector, an element array comprising a first processing element generating a first fixed point output element based on the first fixed point quantization scaled input vector and first plurality of quantization sign bits, and a second processing element generating a second fixed point output element based on the first fixed point quantization scaled input vector and second plurality of quantization sign bits, and a second data type converter generating and outputting first and second output elements by converting data types of the first and second fixed point output elements.

公开(公告)号：US20250117257A1

公开(公告)日：2025-04-10

申请号：US18830998

申请日：2024-09-11

Applicant: Samsung Electronics Co., Ltd.

Inventor： Byungmin AHN ,  Hong Rak SON ,  Dong-Min SHIN ,  Dae-Yeol YANG ,  JongYoon YOON ,  Jae Hun JANG ,  Se Jung KWON ,  Byeongwook KIM ,  Baeseong PARK ,  Dongsoo LEE

IPC: G06F9/50 ,  H03M13/00 ,  H03M13/13

Abstract: Disclosed is an accelerator device which includes an interface circuit that communicates with an external device, a memory that stores first data received through the interface circuit, a polar encoder that performs polar encoding with respect to the first data provided from the memory and to output a result of the polar encoding as second data, and an accelerator core that loads the second data. The first data are compressed weight data, the second data are decompressed weight data, the accelerator core is configured to perform machine learning-based inference based on the second data, and the first data are variable in length.