-
公开(公告)号：WO2022248891A1
公开(公告)日：2022-12-01
申请号：PCT/GB2022/051374
申请日：2022-05-27
Applicant: IMPERIAL COLLEGE INNOVATIONS LTD
Inventor： JANKOWSKI, Mikolaj , TUNG, Tze-Yan , KURKA, David Burth , GUNDUZ, Deniz
Abstract: A communication system is disclosed for conveying data from an information source across a communication channel using joint source and channel coding. The communication system comprises a transmitter having an encoder neural network for encoding input data from an information source direct to values in a signal space for modulating a carrier signal for transmission of a transformed version of the input data across a communication channel. The communication system also comprises a receiver having a decoder neural network for decoding a noise-affected version of the vector signal values transmitted over the noisy communication channel direct to a reconstructed representation of the input data provided from the information source. The transmitter further includes a soft quantization module and a hard quantization module for assigning the output of the encoder to signal values in a predetermined finite set S of symbols of a predefined alphabet of carrier modulation signal values transmittable by the transmitter over the communication channel. A method for training the encoder and decoder neural networks is disclosed that comprises, for input-output pairs of a set of training data from the information source passed to the encoder neural network, determining an objective function characterising a reconstruction error between the input-output pairs of training data, consisting of information source samples, which are passed to the encoder neural network, and the corresponding reconstructions at the decoder output layer of the decoder, using an appropriate optimization algorithm operating on the objective function. An appropriate optimisation algorithm, such as a gradient descent algorithm, is then used on the objective function and the connecting node weights of the hidden layers of the encoder neural network and decoder neural network are updated, to seek to minimise the objective function.
-
22.发明申请
公开(公告)号：WO2022245823A1
公开(公告)日：2022-11-24
申请号：PCT/US2022/029633
申请日：2022-05-17
Applicant: INSURANCE SERVICES OFFICE, INC.
Abstract: Computer vision systems and methods for determining structure features from point cloud data using neural networks are provided. The system obtains point cloud data of a structure or a property parcel having a structure present therein from a database. The system can preprocess the obtained point cloud data to generate another point cloud or 3D representation derived from the point cloud data by spatial cropping and/or transformation, down sampling, up sampling, and filtering. The system can also preprocess point features to generate and/or obtain any new features thereof. Then, the system extracts a structure and/or feature of the structure from the point cloud data utilizing one or more neural networks. The system determines at least one attribute of the extracted structure and/or feature of the structure utilizing the one or more neural networks.
-
公开(公告)号：WO2022245775A1
公开(公告)日：2022-11-24
申请号：PCT/US2022/029562
申请日：2022-05-17
Applicant: LOOP COMMERCE, INC.
Inventor： SAUNKEAH, Gabriel , SIMPSON, Paul
Abstract: Systems and methods for predictive gift subscriptions are provided. A gift subscription service may store a gift subscription and a profile associated with a gift recipient. The gift subscription service may then identify a time to send a gift to the gift recipient and associate a computational model with the gift recipient and the profile. When the gift subscription service determines a change in the profile, the gift subscription service may then update the computational model associated with the gift recipient in real-time by applying a machine learning algorithm to the change in the profile and use the updated computation model to formulate a prediction for the gift recipient that may then be used to identify a set of gifts to present to the gift recipient. The gift subscription service may then present the set of gifts to the gift recipient.
-
公开(公告)号：WO2022234032A2
公开(公告)日：2022-11-10
申请号：PCT/EP2022/062183
申请日：2022-05-05
Applicant: NOVO NORDISK A/S
Inventor： BENGTSSON, Henrik , ARADÓTTIR, Tina, Björk , ENGELL, Sarah, Ellinor , MAHMOUDI, Zeinab , RYDE, Thomas, Emil , LIANG, Bo
IPC: G16H20/17 , A61B5/00 , A61M5/00 , G06N3/00 , G06N20/00 , G16H50/20 , G16H50/30 , A61B5/145 , A61M5/172 , A61B5/14532 , A61B5/7221 , A61B5/7267
Abstract: A method for estimating a fasting blood glucose (FBG) value for a subject, comprising the steps of obtaining a first data set representing a blood glucose history (BGH), obtaining a second data set representing a basal insulin injection history (IH) of the subject, and based on BGH and IH and using a mathematical model: calculating a predicted FBG for the time t+1, where t+1 is the current time. The method comprises the further steps of obtaining at time t+1 a current measured FBG from the subject, and based on the predicted FBG and the current measured FBG: calculating an updated estimated FBG.
-
公开(公告)号：WO2022232404A1
公开(公告)日：2022-11-03
申请号：PCT/US2022/026751
申请日：2022-04-28
Applicant: L'OREAL , KOSECOFF, David B.
Inventor： KOSECOFF, David B.
Abstract: Systems and methods for application of cosmetic designs are described. In an embodiment, the system includes a visible light mirror, comprising a first portion being at least partially transparent to visible light; a camera, optically coupled with the visible light mirror to receive visible light via the first portion; an illumination source, physically coupled with the visible light mirror and configured to emit a plurality of discrete wavelength channels; and a computer system, electronically coupled with the camera and the illumination source, and comprising one or more processors and a non-transitory computer readable storage medium storing instructions that, when executed by the one or more processors, cause the one or more processors to perform one or more methods of the present disclosure.
-
Patent Agency Ranking
公开(公告)号：WO2022231318A1
公开(公告)日：2022-11-03
申请号：PCT/KR2022/006052
申请日：2022-04-27
Applicant: 주식회사 실리콘큐브
Inventor： 임종윤
Abstract: 본 발명의 일실시예는 입차 차량들의 번호판 이미지를 분석하여 상기 입차 차량 별로 입차 차량번호 분석 값 및 상기 분석에 대한 분석 신뢰도 값을 생성하는 입차 관리 단말; 각 상기 입차 차량의 상기 번호판 이미지, 상기 입차 차량번호 분석 값 및 상기 분석 신뢰도 값을 수신하고, 상기 분석 신뢰도 값이 임계치 미만인 입차 차량의 번호판 이미지를 재분석하여 입차 차량번호 재분석 값을 생성하는 서버; 및 출차 차량의 번호판 이미지를 분석하여 출차 차량번호 분석 값을 획득하는 출차 관리 단말을 포함하고, 상기 출차 관리 단말은 상기 입차 차량번호 분석 값 및 상기 입차 차량번호 재분석 값 중 적어도 하나가 상기 출차 차량번호 분석 값과 매칭되는 입차 차량을 상기 출차 차량으로 판단하는 것인 차량 출입 관리 시스템을 제공한다.
-
公开(公告)号：WO2022223941A1
公开(公告)日：2022-10-27
申请号：PCT/GB2022/050764
申请日：2022-03-28
Applicant: JIVA.AI LIMITED
Inventor： PATEL, Manish Ambritbhai
Abstract: Methods, apparatus, and systems are provided for fusing or integrating at least two agent model(s) for modelling a complex system, each agent model comprising: a plurality of agent system (AS) node(s), wherein each of the AS node(s) comprise a plurality of agent units (AUs), and a set of AS rules governing the plurality of AUs, each AU of the plurality of AUs is connected to at least one other AU of the plurality of AUs, an input layer comprising a set of AS nodes of the plurality of AS node(s) an output layer comprising at least one AS node of the plurality of AS node(s) and one or more intermediate layer(s). Each of the intermediate layer(s) comprising another set of AS node(s) of the plurality of AS node(s). Each agent model is trained to model one or more portion(s) of the complex system using a corresponding labelled training dataset. Each agent model being adapted, during training, to form: an agent rule base comprising one or more sets of AS rules, and an agent network state comprising data representative of the interconnections between the AS nodes of the input, output and intermediate layer(s), wherein the agent rule base and agent network state are generated during training and configured for modelling said portion(s) of the complex system. The method comprising: determining an intersecting rule set between the agent rule bases of at least a first trained agent model and a second trained agent model, merging said at least first and second trained agent models to form an integrated agent model based on combining those one or more layer(s), AS node(s), and/or AU(s) of the first and second trained agent models that correspond to the intersecting rule set, and updating the integrated agent model based on one or more validation and training labelled datasets associated with each of the at least first and second trained agent model(s) until the integrated model is validly trained.
-
公开(公告)号：WO2022207443A1
公开(公告)日：2022-10-06
申请号：PCT/EP2022/057746
申请日：2022-03-24
Applicant: BAYER AKTIENGESELLSCHAFT
Inventor： LENGA, Matthias , PURTORAB, Marvin , RAMOS DOS SANTOS, Thiago , HOOGE, Jens , CORONA, Veronica
Abstract: The present invention relates to the technical field of machine learning. Subject matter of the present invention is a novel approach for training a neural network and the use of this approach for the processing of (medical) images.
-
公开(公告)号：WO2022203123A1
公开(公告)日：2022-09-29
申请号：PCT/KR2021/008014
申请日：2021-06-25
Applicant: 주식회사 트랜스버스
Abstract: 캐릭터를 활용한 인공지능 자연어 처리 기반의 화상교육 콘텐츠 제공 방법 및 장치를 개시한다. 본 발명의 실시예에 따른 화상교육 콘텐츠 제공 장치는, 외부 서버로부터 적어도 하나 이상의 참가자의 화상교육 서비스 접속을 확인하는 참가자 확인부; 상기 적어도 하나의 참가자 각각에 대한 영상 및 음성 데이터를 획득하여 참가자 발언 정보를 수집하는 참가자 정보 수집부; 상기 참가자 발언 정보를 발언 텍스트로 변환하여 발언 분석 정보를 생성하는 발언 변환 처리부; 및 상기 발언 분석 정보를 기반으로 캐릭터를 생성하고, 상기 캐릭터를 활용한 화상교육 콘텐츠를 외부 서버를 통해 참가자 단말로 제공하는 캐릭터 형성 처리부를 포함할 수 있다.
-
公开(公告)号：WO2022197766A1
公开(公告)日：2022-09-22
申请号：PCT/US2022/020496
申请日：2022-03-16
Applicant: CAP PATROL
Inventor： THURNER, George
Abstract: Electronic content, including that obtained via software applications executing on client devices, pertaining to golfers' handicap indices and scoring is automatically obtained by golf clubs and other courses. Such player data can be electronically monitored and analyzed in real-time in accordance with expected score qualities and guidelines in the sport. In particular, one or more algorithms may be routinely executed to look for various data point indicators which may mean a player is manipulating his or her handicap index. The systems and methods can objectively generate an overall score for every participating player, based at least in part on deduction of points for data determined abnormal based on expected qualities. Should player's overall score reach a predetermined flagging level, the systems and methods automatically adjust the player's handicap to promote fair play. Machine learning schema can assist in tasks such as scoring analyses and automated handicap adjustments.
-
-
-
-
-
-
-
-
-
Search Results
1343
records
(took 0.031 seconds)
