US 11,853,888 B2
Method and apparatus with neural network convolution operations
Joonho Song, Hwaseong-si (KR); Namjoon Kim, Anyang-si (KR); Sehwan Lee, Suwon-si (KR); and Deokjin Joo, Seongnam-si (KR)
Assigned to Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Dec. 28, 2022, as Appl. No. 18/089,696.
Application 18/089,696 is a continuation of application No. 16/704,290, filed on Dec. 5, 2019, granted, now 11,568,243.
Claims priority of application No. 10-2018-0160342 (KR), filed on Dec. 12, 2018.
Prior Publication US 2023/0169340 A1, Jun. 1, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/08 (2023.01); G06F 7/74 (2006.01); G06N 20/00 (2019.01)
CPC G06N 3/08 (2013.01) [G06F 7/74 (2013.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A processor-implemented method of performing convolution operations in a neural network, the method comprising:
generating a plurality of first sub-bit groups and a plurality of second sub-bit groups, respectively from at least one pixel value of an input feature map and at least one predetermined weight;
performing a convolution operation on a first pair that includes a first sub-bit group including a most significant bit (MSB) of the at least one pixel value and a second sub-bit group including an MSB of the at least one predetermined weight; and
based on a result of the convolution operation on the first pair, determining whether to perform the convolution operations of the remaining pairs.