US 12,423,057 B2
Method and apparatus with deep learning operations with adder tree structure
Hyung-Dal Kwon, Hwaseong-si (KR); Ho Young Kim, Seoul (KR); Hanmin Park, Hwaseong-si (KR); Jaehyeong Sim, Hwaseong-si (KR); Seung Wook Lee, Suwon-si (KR); and Jae-Eon Jo, Suwon-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Jun. 24, 2021, as Appl. No. 17/356,771.
Claims priority of application No. 10-2020-0159177 (KR), filed on Nov. 24, 2020.
Prior Publication US 2022/0164164 A1, May 26, 2022
Int. Cl. G06F 7/544 (2006.01); G06F 7/50 (2006.01)
CPC G06F 7/5443 (2013.01) [G06F 7/50 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus with deep learning, the apparatus comprising:
a systolic adder tree including adder trees connected in row and column directions;
an input multiplexer connected to an input register of at least one of the adder trees and configured to determine column directional data movement between the adder trees based on operation modes; and
a control circuit configured to determine an operation mode for the input multiplexer, among the operation modes, based on either one or both of an operating frequency and an operating voltage of the apparatus.