公开(公告)号：US20200326852A1

公开(公告)日：2020-10-15

申请号：US16674919

申请日：2019-11-05

Applicant: Korea Advanced Institute of Science and Technology

Inventor： Se-Bum Paik ,  Hyeonsu Lee ,  Youngjin Park

IPC: G06F3/06 ,  G06N3/04 ,  G06N3/08

Abstract: Provided are a storage device using a neural network and an operating method of the storage device for automatic redistribution of information and variable storage capacity based on accuracy-storage capacity tradeoff that may learn input information using the neural network and may store the learned information. The neural network may include a plurality of input neurons and a plurality of output neurons; at least one stable synapse configured to connect at least one of the input neurons and at least one of the output neurons, respectively; and at least one flexible synapse configured to connect at least one remaining of the input neurons and at least one remaining of the output neurons, respectively.

公开(公告)号：US10805931B2

公开(公告)日：2020-10-13

申请号：US16286844

申请日：2019-02-27

Applicant: Korea Advanced Institute of Science and Technology

Inventor： Dong Ho Cho ,  Dong Jin Ji ,  Hyuk Joon Lee

IPC: H04W72/04 ,  H04W28/16 ,  G06N3/02

Abstract: Disclosed is a reinforcement learning-based resource allocation method for a wireless backhaul network, which is performed by a resource allocation apparatus. The method includes estimating locations of a plurality of base stations on the basis of channel state information (CSI) measured by the plurality of base stations; and allocating resources of the wireless backhaul network to the plurality of base stations using a reinforcement learning neural network having the locations as an input.

公开(公告)号：US10803814B2

公开(公告)日：2020-10-13

申请号：US15660416

申请日：2017-07-26

Applicant: Korea Advanced Institute of Science And Technology

Inventor： Jun-Bo Yoon ,  Keun Seo Lim

IPC: G09G3/34 ,  G09G3/3208

Abstract: A display apparatus may be provided that includes: a transparent display panel; a blind panel which is disposed adjacent to the transparent display panel and includes a plurality of cells that are individually drivable; and a controller which changes an operation mode through an on/off of the transparent display panel and a selective drive of a cell included in the blind panel. As a result, the display apparatus according to the embodiment of the present invention is a transparent display apparatus using the OLED. The display apparatus is able to operate without the external environmental constraints and to operate in various modes including the display function.

公开(公告)号：US20200319516A1

公开(公告)日：2020-10-08

申请号：US16909858

申请日：2020-06-23

Applicant: SAMSUNG DISPLAY CO., LTD. ,  KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： Kyoung Won PARK ,  Baek Hee LEE ,  Min Ki NAM ,  Hyeok Jin LEE ,  Keun Chan OH ,  Jae Jin LYU ,  Baek Kyun JEON ,  Sungjun KOH ,  Doh Chang LEE

IPC: G02F1/13357 ,  G02F1/1335

Abstract: A quantum dot, a color conversion panel, and a display device, the quantum dot including a core; and a shell layer positioned outside of the core, wherein at least one of the core and the shell layer is doped with aluminum, silicon, titanium, magnesium, or zinc, and the core includes a Group III-V compound.

公开(公告)号：US20200306955A1

公开(公告)日：2020-10-01

申请号：US16804651

申请日：2020-02-28

Applicant: Korea Advanced Institute of Science and Technology

Inventor： Jung KIM ,  Hwayeong JEONG

IPC: B25J9/10

Abstract: Various embodiments provide a variable stiffness soft actuator inspired by an ossicle structure of echinoderm and a robot apparatus including the same. According to various embodiments, the soft actuate includes a plurality of ossicle elements arranged in a specific structure, wherein an interval between the plurality of ossicle elements is maintained or reduced depending on vacuum generation to change the stiffness of the soft actuator.

公开(公告)号：US10790711B2

公开(公告)日：2020-09-29

申请号：US16225781

申请日：2018-12-19

Applicant: KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY ,  UNIVERSITY-INDUSTRY COOPERATION GROUP OF KYUNG HEE UNIVERSITY

Inventor： Joung Ho Kim ,  Bom Son Lee ,  Dong Hyun Kim

IPC: H02J50/90 ,  H02J50/12 ,  H01F27/36 ,  H01F38/14 ,  H01F27/28 ,  H02J7/02 ,  H01F27/255 ,  H01F3/10

Abstract: Disclosed are a magnetic field generating apparatus and a method of generating a magnetic field using the same. A magnetic field generating apparatus according to an embodiment of the present disclosure includes a coil part configured to generate a magnetic field transmitted to the target to which wireless power is to be supplied; a first ferrite member configured to extend in the vicinity of the coil part a bar shape and penetrate an inner peripheral surface of the coil part; and a second ferrite member configured to extend in a dome shape in the vicinity of the coil part, surround the coil part, include an opening formed along the magnetic field transmission path.

公开(公告)号：US20200304300A1

公开(公告)日：2020-09-24

申请号：US16614404

申请日：2018-04-13

Applicant: Korea Advanced Institute of Science and Technology

Inventor： June Koo Rhee ,  Il Hwan Park ,  Kyong Chun Lim ,  Jun Sang Oh ,  Yong Seen Kim

IPC: H04L9/08 ,  H04B10/70 ,  H04B10/63

Abstract: The present invention, which is used for continuous variable quantum key distribution (CVQKD) with asynchronous local oscillators, relates to a system for performing a phase compensation of a scheme of using LO phase estimation and feedback at a receiver (Bob) using a pilot signal from a transmitter (Alice) and a scheme of measuring quantum state data using an LO having a predetermined phase at the receiver (Bob) and estimating and feeding back an LO phase through multi-dimensional reconciliation (MDR).

公开(公告)号：US20200299702A1

公开(公告)日：2020-09-24

申请号：US16811024

申请日：2020-03-06

Applicant: KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： Jung Hoon AHN ,  Hyunjong BYUN ,  Jiyeon PARK

IPC: C12N15/70 ,  C12N15/78 ,  C07K14/21 ,  C12N9/20

Abstract: The present invention provides a method for effective extracellular secretion of a target protein, by preparing a fusion protein connecting LARDS to the target protein and having pI lowered by adjusting the overall charge of target protein, and by using ABC transporter of a bacterial type 1 secretion system (T1SS). The method can allows a protein be produced at a large amount simply and effectively without a separate purification process.

公开(公告)号：US20200290016A1

公开(公告)日：2020-09-17

申请号：US16700571

申请日：2019-12-02

Applicant: Saudi Arabian Oil Company ,  Korea Advanced Institute of Science and Technology

Inventor： Yuguo WANG ,  Rashid M. OTHMAN ,  Minkee CHOI ,  Seokin CHOI

IPC: B01J20/30 ,  B01D53/02 ,  B01J20/18 ,  B01J20/28

Abstract: Provided here are adsorbent compositions containing polyvinyl alcohol-bonded pellets of zeolite templated carbon. Also provided here are methods of producing adsorbent compositions by forming an aqueous mixture containing a binder, water, and zeolite-templated carbon; subjecting the aqueous mixture to a drying process to remove the water and form a dry mixture of the binder and the zeolite-templated carbon, and compacting the dry mixture of the binder and the zeolite-templated carbon to form the binder-bonded pellets of the zeolite templated carbon.

公开(公告)号：US20200285938A1

公开(公告)日：2020-09-10

申请号：US16645344

申请日：2018-09-04

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY

Inventor： Jun-mo KIM ,  Byung-ju KIM ,  Joo-chang KIM ,  Ye-gang LEE ,  Min-seok PARK ,  Ju-seung YUN ,  Jae-young LEE ,  Dong-gyu JOO

IPC: G06N3/04 ,  G06T7/136 ,  G06K9/62

Abstract: The present disclosure provides a neural network training device for recognizing a class of an object included in an image based on an artificial intelligence (AI) system and an application thereof, the neural network training method including: acquiring, by using a first learning network model trained based on source training images respectively included in at least one class, feature information of a query image included in a class different from the at least one class; obtaining a generated image from the feature information of the query image by using a second learning network model acquiring feature information of the obtained generated image by using the first learning network model; and updating weights of layers respectively included in the first and second learning network models, based on a difference between the feature information of the query image and the feature information of the generated image and on a difference between the query image and the generated image.