公开(公告)号：EP4443339A1

公开(公告)日：2024-10-09

申请号：EP22922194.0

申请日：2022-12-22

申请人： OMRON Corporation

发明人： YANAGI, Rintaro ,  HASHIMOTO, Atsushi ,  CHIBA, Naoya ,  USHIKU, Yoshitaka

IPC分类号： G06N3/09 ,  G06N3/04

CPC分类号： G06N3/04 ,  G06N20/00 ,  G06N3/09

摘要： The model generation device performs machine learning for an inference model that includes a preprocessing module and a graph inference module. The preprocessing module includes a feature extractor and a selection module. The feature extractor calculates a feature value of each element belonging to one of a plurality of sets included in the input graph. The selection module selects one or more edges extending from each element as a starting point based on the calculated feature value of each element, and generates graph information, indicating the calculated feature value of each element and an edge selection result for each set. The graph inference module is configured to be differentiable and infers a solution to a task for the input graph from the generated graph information for each set.

公开(公告)号：EP4432233A1

公开(公告)日：2024-09-18

申请号：EP22893179.6

申请日：2022-11-09

申请人： Samsung Electronics Co., Ltd.

发明人： DINH, Quockhanh ,  CHOI, Kwangpyo

IPC分类号： G06T9/00 ,  H04N19/51 ,  H04N19/147 ,  H04N19/184 ,  H04N19/44 ,  G06N3/08 ,  G06N3/04 ,  H04N19/573 ,  H04N19/159

CPC分类号： G06N3/04 ,  G06N3/08 ,  G06T9/00 ,  H04N19/147 ,  H04N19/159 ,  H04N19/184 ,  H04N19/44 ,  H04N19/573 ,  H04N19/51

摘要： An image decoding method according to an embodiment may comprise the steps of: obtaining a current optical flow, and correction data generated based on a current prediction image and a current image, from a bitstream; obtaining the current prediction image on the basis of the current optical flow and a previous reconstruction image; obtaining feature data of the current prediction image by applying the current prediction image to a neural network-based prediction image encoder; and obtaining a current reconstruction image corresponding to the current image by applying the correction data and the feature data of the current prediction image to a neural network-based image decoder.

公开(公告)号：EP4430524A1

公开(公告)日：2024-09-18

申请号：EP22891242.4

申请日：2022-11-10

申请人： Deeplite Inc.

发明人： FERIANC, Martin ,  SANKARAN, Anush ,  MASTROPIETRO, Olivier ,  SABOORI, Ehsan ,  SAWYER, Davis Mangan

IPC分类号： G06N3/08

CPC分类号： G06N3/082 ,  G06N3/04

公开(公告)号：EP4421688A1

公开(公告)日：2024-08-28

申请号：EP22938660.2

申请日：2022-11-16

申请人： Postech Academy-Industry Foundation

发明人： HWANG, Hyun Sang ,  CHOI, Woo Seok

IPC分类号： G06N3/08 ,  G06N3/063 ,  G06N3/04 ,  G06N5/04

CPC分类号： G06N5/04 ,  G06N3/04 ,  G06N3/063 ,  G06N3/08

摘要： A learning apparatus and method for implementing an edge device using a resistive element and an analysis apparatus and method using the same are disclosed. The learning apparatus for implementing an edge device using a resistive element according to an embodiment of the present application may include a first learning unit determining a weight of an artificial neural network through learning based on first training data and reflecting the determined weight in a first resistive element, and a second learning unit updating the weight of the artificial neural network through learning based on second training data collected through the device and reflecting the updated weight in the second resistive element.

公开(公告)号：EP4412229A1

公开(公告)日：2024-08-07

申请号：EP22916566.7

申请日：2022-12-15

申请人： Samsung Electronics Co., Ltd.

发明人： KWON, Jungmin ,  PARK, Chiyoun ,  KANG, Byeongcheol ,  PARK, Hyunseo ,  SOHN, Youngchul ,  SONG, Juyong

IPC分类号： H04N21/485 ,  H04N21/466 ,  G06F3/04842 ,  G06F3/0481 ,  G06N3/08 ,  G06N3/04 ,  G06F3/0482

CPC分类号： G06F3/04842 ,  G06F3/0481 ,  G06F3/0482 ,  G06N3/04 ,  G06N3/08 ,  H04N21/466 ,  H04N21/485

摘要： A method for analyzing an application screen is disclosed. The method comprises the steps of: generating a plurality of links for a plurality of user interface elements included in an application screen; generating a user interface map for each of one or more primitive actions, which are user inputs for searching for an application screen on the basis of the plurality of links; and identifying the location of a focus indicating a user interface element that is to interact with a user from among the plurality of user interface elements, wherein the user interface map includes a path along which the location of the focus moves between the plurality of user interface elements by means of the one or more primitive actions.

公开(公告)号：EP4406481A2

公开(公告)日：2024-07-31

申请号：EP22873175.8

申请日：2022-09-21

申请人： Medical AI Co., Ltd.

发明人： KWON, Joonmyoung ,  JANG, Jonghwan

IPC分类号： A61B5/349 ,  A61B5/00 ,  G16H50/20 ,  G16H50/50 ,  G16H50/70 ,  G06N3/08 ,  G06N3/04

CPC分类号： A61B5/00 ,  A61B5/349 ,  G06N3/04 ,  G06N3/08 ,  G16H50/20 ,  G16H50/50 ,  G16H50/70

摘要： According to an embodiment of the present disclosure, there are disclosed a method, program and device for interpreting medical data based on explainable artificial intelligence, which are performed by a computing device. The method may include: training a first neural network model to estimate a positive or negative for a disease with respect to first medical data based on the first medical data; and training a second neural network model configured to transform features of second medical data so that a positive or negative for the disease with respect to the second medical data is estimated to be opposite by using the trained first neural network model.

公开(公告)号：EP4402791A1

公开(公告)日：2024-07-24

申请号：EP22803322.1

申请日：2022-10-24

申请人： Siemens Aktiengesellschaft

发明人： HEIN, Daniel ,  LIMMER, Steffen ,  MONDAL, Gopal ,  NIELEBOCK, Sebastian ,  STERZING, Volkmar

IPC分类号： H02M7/5387 ,  H02M7/5395 ,  H02M1/00 ,  H02P27/08 ,  G05B13/02 ,  G06N3/08

CPC分类号： H02M7/53873 ,  H02M7/5395 ,  H02M1/0012 ,  H02P27/08 ,  H02P27/085 ,  G06N3/04 ,  G06N3/09

公开(公告)号：EP4068033B1

公开(公告)日：2024-07-10

申请号：EP21305408.3

申请日：2021-03-31

IPC分类号： G05D1/48 ,  G05D1/689 ,  G05D1/00 ,  G06F11/36 ,  B25J9/16 ,  G06N3/02 ,  G06N3/084 ,  G06N3/04 ,  G05B13/02 ,  G05B17/02 ,  G05D101/15 ,  G05D107/70 ,  G05D109/10

CPC分类号： G05B13/027 ,  G05B17/02 ,  G06N3/084 ,  G06N3/04 ,  G06F30/17 ,  G06F30/20 ,  G06F30/27 ,  G05D2109/1020240101 ,  G05D1/48 ,  G05D1/6895 ,  G05D2101/1520240101 ,  G05D2107/7020240101

公开(公告)号：EP4394726A1

公开(公告)日：2024-07-03

申请号：EP22909548.4

申请日：2022-11-04

申请人： Tencent Technology (Shenzhen) Company Limited

发明人： ZHAO, Yu ,  LIN, Zhenyu ,  YAO, Jianhua

IPC分类号： G06V10/764 ,  G06V10/26 ,  G06V10/44 ,  G06V10/80 ,  G06V10/82 ,  G06F18/40 ,  G06N3/04

CPC分类号： G06V10/82 ,  G06N3/04 ,  G06N3/08 ,  G06V10/451

摘要： Embodiments of the present application relate to the technical field of computers, and disclosed are an image classification method and apparatus, a computer device, and a storage medium. The method comprises: obtaining image features of a pathological image to be classified; for each scale in a plurality of scales, extracting local features corresponding to the scale from the image features; performing splicing processing according to the local features corresponding to each scale to obtain a spliced image feature; and classifying the spliced image feature to obtain a category to which the pathological image belongs. According to the method provided by the embodiments of the present application, local features corresponding to different scales contain different information, so that the finally obtained spliced image feature contains feature information corresponding to different scales, and the feature information of the spliced image feature is enriched; and the category to which the pathological image belongs is determined on the basis of the spliced image feature, so that the accuracy of the category is ensured.

公开(公告)号：EP4384962A1

公开(公告)日：2024-06-19

申请号：EP21953452.6

申请日：2021-11-30

申请人： Imemori Technologies Private Limited

发明人： ANAOKAR, Malhar ,  PRASAD, Akshat

IPC分类号： G06Q10/06 ,  G06Q10/10

CPC分类号： G06Q10/10 ,  G06F40/30 ,  G06F40/20 ,  G06N3/04 ,  G06N3/08