公开(公告)号：US20230008124A1

公开(公告)日：2023-01-12

申请号：US17784856

申请日：2020-12-11

Applicant: Korea Electronics Technology Institute ,  INDUSTRY-UNIVERSITY COOPERATION FOUNDATION KOAEA AEROSPACE UNIVERSITY

Inventor： Byeong Ho CHOI ,  Sang Seol LEE ,  Sung Joon JANG ,  Sung Jei KIM ,  Jae Gon KIM ,  Hyeon Cheol MOON

IPC: G06N3/04

Abstract: Disclosed herein are a method and apparatus for encoding/decoding a deep neural network. According to the present disclosure, the method for decoding a deep neural network may include: in a plurality of layers of the deep neural network, entropy decoding quantization information for a current layer; performing dequantization on the current layer; and obtaining a plurality of layers of the deep neural network. At least one of global quantization and local quantization is performed on the current layer.

公开(公告)号：US20230010859A1

公开(公告)日：2023-01-12

申请号：US17784862

申请日：2020-11-30

Applicant: Korea Electronics Technology Institute

Inventor： Byeong Ho CHOI ,  Sang Seol LEE ,  Sung Joon JANG ,  Sung Jei KIM

IPC: G06N3/08

Abstract: Disclosed herein are a method and apparatus for encoding/decoding a deep learning network. According to an embodiment, the method for decoding a deep learning network may include decoding network header information regarding the deep learning network; decoding layer header information regarding a plurality of layers in the deep learning network; decoding layer data information regarding specific information of the plurality of layers; and obtaining the deep learning network and a plurality of layers in the deep learning network, and the layer header information includes layer distinction information associated with distinguishing the plurality of layers.