专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20230122985A1 SINGLE-FRAME FRINGE PATTERN ANALYSIS METHOD BASED ON MULTI-SCALE GENERATIVE ADVERSARIAL NETWORK 有权
公开(公告)号：US20230122985A1
公开(公告)日：2023-04-20
申请号：US17909780
申请日：2020-08-27
申请人： Nanjing University of Science and Technology
发明人： Shijie FENG , Qian CHEN , Chao ZUO , Yuzhen ZHANG , Jiasong SUN , Yan HU , Wei YIN , Jiaming QIAN
IPC分类号： G06N3/094 , G06N3/0475 , G06N3/048 , G06V10/44
摘要： The invention discloses a single-frame fringe pattern analysis method based on multi-scale generative adversarial network. A multi-scale generative adversarial neural network model is constructed and a comprehensive loss function is applied. Next, training data are collected to train the multi-scale generative adversarial network. During the prediction, a fringe pattern is fed into the trained multi-scale network where the generator outputs the sine term, cosine term, and the modulation image of the input pattern. Finally, the arctangent function is applied to compute the phase. When the network is trained, the parameters of the network do not need to manually tune during the calculation. Since the input of the neural network is only a single fringe pattern, the invention provides an efficient and high-precision phase calculation method for moving objects.

2. 发明申请

US20230095977A1 INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM 有权
公开(公告)号：US20230095977A1
公开(公告)日：2023-03-30
申请号：US17802699
申请日：2020-03-19
申请人： Sony Interactive Entertainment Inc.
发明人： Yoshikazu ONUKI
IPC分类号： G06T7/20 , G06N3/0475 , G06N3/094 , G06N3/088
摘要： There is provided an information processing apparatus including a generator building section configured to build a generator for generating image data of a first domain from image data of a second domain and for generating the image data of the second domain from the image data of the first domain by use of cycle GAN (generative adversarial networks). The first domain is defined by image data of a plurality of channels including at least two time series images each including a first object and a second object, each of the objects having a tendency to move differently. The second domain is defined by the image data of a plurality of the channels including at least the two time series images each including the first object and excluding the second object.

3. 发明申请

US20230134186A1 MACHINE LEARNING DATA GENERATION DEVICE, MACHINE LEARNING MODEL GENERATION METHOD, AND STORAGE MEDIUM 有权
公开(公告)号：US20230134186A1
公开(公告)日：2023-05-04
申请号：US18088766
申请日：2022-12-26
申请人： Kabushiki Kaisha Yaskawa Denki
发明人： Ryohei SUZUKI , Tsuyoshi Yokoya , Ryo Masumura , Hiroki Tachikake
IPC分类号： G06N3/094
摘要： Provided a machine learning data generation device including: at least one processor; and at least one memory device that stores a plurality of instructions which, when executed by the at least one processor, causes the at least one processor to execute: acquiring , in association with a predetermined label actual time series information; executing physical simulation of generating a plurality of pieces of virtual time series information; identifying parameter values based on the plurality of pieces of virtual time series information and the actual time series information, and to associate the identified parameter values with the label; generating a new parameter value and the label based on the identified parameter values; generating virtual time series information corresponding to a new internal state by executing physical simulation through use of the new parameter value; and generating new machine learning data.

4. 发明申请

US20230116764A1 APPARATUS AND METHOD FOR PERSONAL IDENTIFICATION BASED ON DEEP NEURAL NETWORK 有权
公开(公告)号：US20230116764A1
公开(公告)日：2023-04-13
申请号：US17964441
申请日：2022-10-12
申请人： RESEARCH & BUSINESS FOUNDATION SUNGKYUNKWAN UNIVERSITY
发明人： Yu Sung KIM , Seung Hwan SHIN , Keun Hong CHAE , Seong Hyun BAN , Seung Hyeon KIM
IPC分类号： G06N3/094
摘要： The present disclosure relates to an apparatus and a method for personal identification based on a deep neural network. According to an exemplary embodiment of the present disclosure, a personal identification method includes receiving a plurality of wireless signals including spatial information and identification information of an object to be identified, by means of a plurality of receivers in different positions, by a wireless signal collecting unit, generating a manipulation signal from the wireless signal by processing the spatial information by means of a first deep neural network model which is trained in advance, by a manipulation signal generating unit, and identifying the object to be identified in a specific space with identification information of the object to be identified of the manipulation signal as an input of the second deep neural network model, by a personal identification processing unit.

5. 发明申请

WO2023084342A1 電気化学セパレータのデータ処理装置及び設計方法审中-公开
公开(公告)号：WO2023084342A1
公开(公告)日：2023-05-19
申请号：PCT/IB2022/060081
申请日：2022-10-20
申请人：ロベルト·ボッシュ·ゲゼルシャフト·ミト•ベシュレンクテル·ハフツング
发明人：吉田　力矢
IPC分类号： G06F30/17 , G06F30/27 , G06N3/0475 , G06N3/094 , H01M50/489 , H01M8/0202 , H01M8/0258 , H01M8/04992
摘要：電気化学電池のセパレータの設計を支援するデータ処理装置(1)は、前記セパレータの設計データを生成する生成部(21)を備える。前記生成部(21)は、異なるバリエションを有する複数の前記設計データを生成する生成器(211)を備える。前記生成器(211)は、初期データの入力に対して、先に生成された前記設計データと異なる前記設計データを出力する学習モデルであり、訓練データの特徴と入力されたデータの特徴との類似性に関して識別する識別器(212)の識別結果に基づいて予め訓練されている。

6. 发明公开

EP4328799A1 A METHOD FOR TRAINING DETERMINISTIC AUTOENCODERS 审中-公开
公开(公告)号：EP4328799A1
公开(公告)日：2024-02-28
申请号：EP22192386.5
申请日：2022-08-26
申请人： Robert Bosch GmbH
发明人： Saseendran, Amrutha , Keuper, Margret , Skubch, Kathrin
IPC分类号： G06N3/0455 , G06N3/088 , G06N3/094 , G06N7/01
摘要： A method (100) for training a deterministic autoencoder (1) for records (4) of measurement data, wherein the autoencoder (1) comprises an encoder (2) that maps a record (4) of measurement data as input to a representation (5) in a latent space that has a lower dimensionality than the input (4) in a deterministic manner and a decoder (3) that reconstructs a record (4#) of measurement data from the representation (5) in a deterministic manner, the method (100) comprising the steps of:
• providing (110) training records (4a) of measurement data;
• determining (120) adversarial records (4b) that do not belong to a distribution and/or domain to which the training records (4a) of measurement data belong;
• processing (130), by the encoder (2), both training records (4a) and adversarial records (4b) into respective representations (5a, 5b);
• reconstructing (140), by the decoder (3), records (4#) of measurement data from the representations (5a, 5b);
• evaluating (150) a loss function (6) that measures
∘ how similar the reconstructed records (4#) of measurement data are to the corresponding training records (4a), respectively adversarial records (4b); and
∘ how similar distributions of the representations (5a, 5b) are to a given prior distribution; and

• optimizing (160) parameters (1a) that characterize the behavior of the autoencoder (1) towards the goal that, when processing further training records (4a) and adversarial records (4b), the value (6a) of the loss function (6) improves.

7. 发明公开

EP4202786A1 ADVERSARIAL SAMPLE PROTECTION FOR MACHINE LEARNING 审中-公开
公开(公告)号：EP4202786A1
公开(公告)日：2023-06-28
申请号：EP22201179.3
申请日：2022-10-12
申请人： INTEL Corporation
发明人： LEVY, Dor , BEN-SHALOM, Omer , KELLERMANN, Raizy , NAYSHTUT, Alex
IPC分类号： G06N3/094 , G06F21/54 , G06F21/57 , G06N3/084
摘要： Adversarial sample protection for machine learning is described. An example of a storage medium includes instructions for initiating processing of examples for training of an inference engine in a system; dynamically selecting a subset of defensive preprocessing methods from a repository of defensive preprocessing methods for a current iteration of processing, wherein a subset of defensive preprocessing methods is selected for each iteration of processing; performing training of the inference engine with a plurality of examples, wherein the training of the inference engine include operation of the selected subset of defensive preprocessing methods; and performing an inference operation with the inference engine, including utilizing the selected subset of preprocessing defenses for the current iteration of processing.

8. 发明申请

WO2023087006A1 DEEP LEARNING FOR MODELING DISEASE PROGRESSION 审中-公开
公开(公告)号：WO2023087006A1
公开(公告)日：2023-05-19
申请号：PCT/US2022/079839
申请日：2022-11-14
申请人： GENENTECH, INC.
发明人： HEJRATI, Seyed Mohammadmohsen , HASHEMIFAR, Somaye Sadat , IRIONDO, Claudia
IPC分类号： G06N3/0464 , G06N3/0455 , G06N3/094
摘要： A method is provided for deep learning for modeling disease progression. The method may include generating, by a machine learning model, a first feature representation based on clinical data associated with a baseline cognitive state of a patient. The method may also include generating, by the machine learning model, a second feature representation based on an image of a brain of the patient. The method may also include generating, by the machine learning model, a set representation by at least fusing the first feature representation and the second feature representation. The method may also include predicting, by the machine learning model, a change in the baseline cognitive state over a time period based at least on the set representation. Related systems and articles of manufacture are also disclosed.

9. 发明申请

WO2023076462A1 USING MULTIPLE DEVICES TO MONITOR PHYSIOLOGICAL DATA 审中-公开
公开(公告)号：WO2023076462A1
公开(公告)日：2023-05-04
申请号：PCT/US2022/048004
申请日：2022-10-27
申请人： MONOVO, LLC
发明人： TAYLOR, Steven , ELLIOT, Kinnear
IPC分类号： A61B5/0205 , A61B5/00 , G06N3/0895 , G06N3/092 , G06N3/094 , G06N3/096 , G06N3/0985
摘要： A method for using a plurality of wearable devices to monitor physiological data includes receiving one or more first sets of physiological data from a first wearable device that is placed at a first location of a wearer body and receiving one or more second sets of physiological data from a second wearable device that is placed at a second location of a wearer body. The method also includes generating one or more overall values of one or more physiological metrics based on the one or more first sets of physiological data and the one or more second sets of physiological data and sending the one or more overall values of the one or more physiological metrics to a cloud service.

10. 发明公开

EP4322058A1 OPTIMIZED DATA STORAGE IN MOBILE NETWORK SCENARIOS 审中-公开
公开(公告)号：EP4322058A1
公开(公告)日：2024-02-14
申请号：EP22189723.4
申请日：2022-08-10
申请人： Nokia Solutions and Networks Oy
发明人： Alberi-Morel, Marie Line , Conte, Alberto , Giust, Fabio , Kaada, Soumeya
IPC分类号： G06N3/045 , G06N3/047 , G06N3/02 , G06N3/0475 , G06N3/098 , G06N3/063 , G06N3/094
摘要： There are provided measures for optimized data storage in mobile network scenarios. Such measures exemplarily comprise transmitting a data storage request requesting storage of data, said data storage request including conveyed information and an indicator indicative of a degree of involvement of generative models in said conveyed information which represents said data, and receiving a data storage acknowledge response.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式