公开(公告)号：US20220320425A1

公开(公告)日：2022-10-06

申请号：US17836435

申请日：2022-06-09

Applicant: Samsung Electronics Co., Ltd. ,  President and Fellows Of Harvard College

Inventor： Minhyun LEE ,  Dovran AMANOV ,  Renjing XU ,  Houk JANG ,  Haeryong KIM ,  Hyeonjin SHIN ,  Yeonchoo CHO ,  Donhee HAM

IPC: H01L45/00 ,  H01L27/24

Abstract: Provided are memristors and neuromorphic devices including the memristors. A memristor includes a lower electrode and an upper electrode that are apart from each other and first and second two-dimensional material layers that are arranged between the lower electrode and the upper electrode and stacked without a chemical bond therebetween.

公开(公告)号：US20220147799A1

公开(公告)日：2022-05-12

申请号：US17500429

申请日：2021-10-13

Applicant: Samsung Electronics Co., Ltd. ,  President and Fellows Of Harvard College

Inventor： Changhyun KIM ,  Houk JANG ,  Henry Julian HINTON ,  Hyeonjin SHIN ,  Minhyun LEE ,  Donhee HAM

IPC: G06N3/04 ,  G06K9/62 ,  G06T7/11

Abstract: Disclosed is a neural computer including an image sensor capable of controlling a photocurrent. The neural computer according to an embodiment includes a preprocessor configured to receive an image and generate a feature map for the received image; a flattening unit configured to transform the feature map generated by the preprocessor into tabular data to provide data output; and an image classifier configured to classify images received through the preprocessor by using the data output by the flattening unit as an input value. The preprocessor includes an optical signal processor configured to receive the image and generate the feature map.

公开(公告)号：US20190013648A1

公开(公告)日：2019-01-10

申请号：US15962600

申请日：2018-04-25

Applicant: Samsung Electronics Co., Ltd. ,  President and Fellows Of Harvard College

Inventor： Jinseong HEO ,  Minhyun LEE ,  Seongjun PARK ,  Philip KIM ,  Hongkun PARK ,  Donhee HAM

IPC: H01S5/30 ,  H01S5/062 ,  H01S5/042 ,  H01S5/323

Abstract: A light-emitting device includes a substrate including a photonic cavity and configured to function as a gate, an active layer including a two-dimensional material, a first conductive contact, and a second conductive contact. The wavelength range of light generated by the light-emitting device may be narrowed based on the photonic cavity being included in the substrate, and the intensity and wavelength range of the generated light may be controlled based on the substrate functioning as a gate.

公开(公告)号：US20230189673A1

公开(公告)日：2023-06-15

申请号：US18167354

申请日：2023-02-10

Applicant: Samsung Electronics Co., Ltd. ,  President and Fellows Of Harvard College

Inventor： Minhyun LEE ,  Houk JANG ,  Donhee HAM ,  Chengye LIU ,  Henry Julian HINTON ,  Haeryong KIM ,  Hyeonjin SHIN

IPC: H10N70/00 ,  H10B63/00

CPC classification number: H10N70/8836 ,  H10B63/84 ,  H10N70/011 ,  H10N70/841 ,  H10N70/8845

Abstract: A nonvolatile memory device includes a resistance switching layer, a gate on the resistance switching layer, a gate oxide layer between the resistance switching layer and the gate, and a source and a drain, spaced apart from each other, on the resistance switching layer. A resistance value of the resistance switching layer is changed based on an illumination of light irradiated onto the resistance switching layer and is maintained as a changed resistance value.

公开(公告)号：US20220147805A1

公开(公告)日：2022-05-12

申请号：US17523977

申请日：2021-11-11

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Donhee HAM ,  Sang Joon KIM

IPC: G06N3/063 ,  G11C11/54

Abstract: Disclosed is an apparatus and method mapping a natural neural network into an electronic neural network device of an electronic device. The method includes constructing a neural network map of a natural neural network based on membrane potentials of a plurality of biological neurons of the natural neural network, where the membrane potentials correspond to at least two different respective forms of membrane potentials, and mapping the neural network map to the electronic neural network device. The constructing of the neural network map and the mapping of the neural network map implement learning of the electronic neural network device. The method may further includes obtaining an input or stimuli, activating the learned electronic neural network device, provided the obtained input or stimuli, to perform neural network operations, and generating a neural network result for the obtained input or stimuli based on a result of the activated learned electronic neural device.

公开(公告)号：US20210151678A1

公开(公告)日：2021-05-20

申请号：US17094121

申请日：2020-11-10

Applicant: Samsung Electronics Co., Ltd. ,  President and Fellows Of Harvard College

Inventor： Minhyun LEE ,  Houk JANG ,  Donhee HAM ,  Chengye LIU ,  Henry HINTON ,  Haeryong KIM ,  Hyeonjin SHIN

IPC: H01L45/00 ,  H01L27/24

Abstract: A nonvolatile memory device includes a resistance switching layer, a gate on the resistance switching layer, a gate oxide layer between the resistance switching layer and the gate, and a source and a drain, spaced apart from each other, on the resistance switching layer. A resistance value of the resistance switching layer is changed based on an illumination of light irradiated onto the resistance switching layer and is maintained as a changed resistance value.