公开(公告)号：WO2023061209A1

公开(公告)日：2023-04-20

申请号：PCT/CN2022/121694

申请日：2022-09-27

申请人： 中兴通讯股份有限公司

发明人： 易哲 ,  黄景丰 ,  郑紫阳 ,  陈晓艳

IPC分类号： G06F11/10 ,  G06N20/10

摘要： 本申请涉及一种内存故障的预测方法、电子设备和计算机可读存储介质。其中内存故障的预测方法包括：获取待测内存的多种日志数据；其中，所述多种日志数据至少包括：内存错误信息地址数据；根据所述多种日志数据进行特征工程构造，得到所述多种日志数据分别对应的特征数据表；对所述多种日志数据分别对应的特征数据表进行拼接，得到特征拼接数据表；根据所述特征拼接数据表和预训练的故障预测模型，得到所述待测内存的故障预测结果；其中，所述故障预测模型根据预先采集的训练数据集训练得到，所述训练数据集中的样本包括多种内存的多种日志数据。

公开(公告)号：WO2023034028A1

公开(公告)日：2023-03-09

申请号：PCT/US2022/040687

申请日：2022-08-18

申请人： SIEMENS AKTIENGESELLSCHAFT ,  SIEMENS CORPORATION

发明人： AKROTIRIANAKIS, Ioannis ,  DEY, Biswadip ,  CHAKRABORTY, Amit

IPC分类号： G06N5/00 ,  G06N20/10 ,  G05D1/00 ,  G06N3/00

摘要： System and method are disclosed for approximating unknown safety constraints during reinforcement learning of an autonomous agent. A controller for directing the autonomous agent includes a reinforcement learning (RL) algorithm configured to define a policy for behavior of the autonomous agent, and a control barrier function (CBF) algorithm configured to calculate a corrected policy that relocates policy states to an edge of a safety region. Iterations of the RL algorithm safely learn an optimal policy where exploration remains within the safety region. CBF algorithm uses standard least squares to derive estimates of coefficients for linear constraints of the safe region. This overcomes inaccurate estimation of safety region constraints caused by one or more noisy observations of constraints received by sensors.

公开(公告)号：WO2023023748A1

公开(公告)日：2023-03-02

申请号：PCT/AU2022/050975

申请日：2022-08-24

申请人： 3P HEALTHCARE PTY LTD

发明人： BEAVER, Paul

IPC分类号： G06K9/62 ,  A61B5/00 ,  A61B5/024 ,  A61B5/145 ,  G16H10/60 ,  G16H20/30 ,  G01N33/50 ,  G06N20/10 ,  G06N3/02 ,  G16H50/20 ,  G16H50/30

摘要： A cardiovascular health assessment model that identifies as well as stratifies individuals at risk of cardiovascular (CV) disease using low cost biomarkers more effectively than traditional CV models. The model allows for better management of 'at risk' or 'high risk' individuals by their medical / healthcare practitioner regarding the efficacy of nutrition, exercise, and lifestyle interventions. Individuals can take responsibility for their own health by having the ability to monitor their own health data, using new digital healthcare technologies connected to a bioinformatics platform, and thereby reduce the risk of a future adverse health event. The model enables practitioners to successfully integrate digital health and bioinformatics into their clinics to further improve the health and well-being of their patients, enhance the performance of their clinics, and ultimately reduce community healthcare costs.

公开(公告)号：WO2023280569A1

公开(公告)日：2023-01-12

申请号：PCT/EP2022/067016

申请日：2022-06-22

申请人： METACLUSTER LT, UAB

发明人： KUKSTA, Andrius ,  GORSKOVAS, Jurijus ,  JURAVICIUS, Martynas

IPC分类号： G06F16/954 ,  G06F16/957 ,  G06F16/986 ,  G06N20/00 ,  G06N20/10 ,  G06N20/20 ,  G06N3/0464 ,  G06N3/09 ,  G06N5/01 ,  G06N5/022

摘要： The current application discloses processor-implemented methods and systems of processing unclassified HTML responses collected in the context of a data collection service, the method comprising, in one embodiment, receiving unclassified HTML documents, isolating elements relevant for category identification, deriving classification attributes from the isolated elements, and applying a Machine Learning-based classification model resulting in HTML data items classified and labelled accordingly. In certain embodiments the Machine Learning model may be a model trained on a pre-created training data set labeled manually or in an automatic fashion.

公开(公告)号：WO2023277480A1

公开(公告)日：2023-01-05

申请号：PCT/KR2022/009122

申请日：2022-06-27

申请人： 한국과학기술원

发明人： 이준구 ,  박시헌 ,  박경덕

IPC分类号： G06N10/00 ,  G06N20/10 ,  B82Y10/00

摘要： 본 발명은 NISQ 컴퓨터 상에서 실현 가능한 데이터 분류 장치에 관한 것으로, 양자 근사 서포트 벡터 머신(QASVM) 알고리즘의 목적함수를 최소화하는 가중치 정보를 산출하는 가중치 산출부; 및 상기 가중치 산출부로부터 획득된 가중치 정보를 이용하여 상기 QASVM 알고리즘의 분류점수를 계산하고, 상기 계산된 분류점수를 기반으로 입력 데이터의 클래스를 분류하는 데이터 분류부를 포함한다.

公开(公告)号：WO2022263897A1

公开(公告)日：2022-12-22

申请号：PCT/IB2021/055382

申请日：2021-06-17

申请人： STUDIO MODERNA BRANDS INTERNATIONAL LIMITED

发明人： RIBICIC, Boris ,  BARADOUSKA, Ieva

IPC分类号： G06Q30/06 ,  G06N3/02 ,  G06N20/10 ,  G06N3/006 ,  G06N3/0454 ,  G06N3/084 ,  G06Q30/0621 ,  G06V40/103 ,  G06V40/168

摘要： A customized mattress may be provided to a user employing a user device. The user is provided with customization options for the mattress. In response to receiving the customization options, the user device sends user input data to a central server system. In addition, generalized consumer proportional dimension data for the user is generated by the user or by an image generation server. Based on the input data and the generalized consumer proportional dimension data, placement of various types of foam springs in ergonomic contoured configuration areas is determined. Each of types of foam springs have a corresponding strength and density rating. The central server system retrieves a mapping of the mattress corresponding to the placement and transmits mapping to the user device. The user may then purchase the customized mattress corresponding to the mapping.

公开(公告)号：WO2022257243A1

公开(公告)日：2022-12-15

申请号：PCT/CN2021/108170

申请日：2021-07-23

申请人： 西安邮电大学 ,  西安碧海蓝天电子信息技术有限公司

发明人： 高强 ,  赵小强

IPC分类号： G01N33/18 ,  G06N3/00 ,  G06N20/10

摘要： 一种水质预警方法及系统，方法包括：获取待测水域的当前生物运动特征数据(101)；将待测水域的当前生物运动特征数据，输入水质预警模型，得到预警结果(102)；水质预警模型是采用差分进化算法和灰狼优化算法对支持向量机模型进行训练得到的；该方法及系统通过差分进化算法和灰狼优化算法对支持向量机模型进行训练得到水质预警模型，使用水质预警模型对待测水域进行水质预警，能够提高水质预警的准确性，具有全局搜索能力强的优点。

公开(公告)号：WO2022253463A1

公开(公告)日：2022-12-08

申请号：PCT/EP2022/025249

申请日：2022-05-27

申请人： ARM LIMITED

发明人： POTTIER, Remy ,  CROXFORD, Daren ,  CARPENTER, Gary, Dale

IPC分类号： G06N3/04 ,  G06N3/08 ,  G06N5/00 ,  G06N7/00 ,  G06N20/10 ,  G06N20/20 ,  G06N5/04

摘要： Briefly, example methods, apparatuses, and/or articles of manufacture are disclosed that may be implemented, in whole or in part, using one or more computing devices to enhance capabilities of peer devices. In an implementation, at least one agent to: identify one or more learnable capabilities enabled by one or more parameters that are accessible via receipt of one or more message at the one or more communication devices from one or more other computing devices; and determine a utility of augmenting at least one of the one or more learning engines with at least one of the one or more learnable capabilities.

公开(公告)号：WO2022247070A1

公开(公告)日：2022-12-01

申请号：PCT/CN2021/119290

申请日：2021-09-18

申请人： 北京工业大学

发明人： 方娟 ,  杨会静 ,  滕自怡 ,  蔡旻

IPC分类号： G06F12/127 ,  G06N20/10 ,  G06F12/0862 ,  G06F12/0811

摘要： 一种面向高性能的适应预取的智能缓存替换策略，在存在硬件预取器的情况下，区分预取和需求请求，利用基于ISVM(Integer Support Vector Machines)的预取预测器对预取访问加载的缓存行进行重引用间隔预测，利用基于ISVM的需求预测器对需求访问加载的缓存行进行重引用间隔预测。输入当前访存的load指令的PC地址和访存历史记录中过去load指令的PC地址，针对预取和需求请求设计不同的ISVM预测器，以请求类型为粒度对加载的缓存行进行重用预测，改善存在预取时缓存行重用预测的准确度，更好的融合了硬件预取和缓存替换带来的性能提升。

公开(公告)号：WO2022244547A1

公开(公告)日：2022-11-24

申请号：PCT/JP2022/017395

申请日：2022-04-08

申请人： 株式会社日立製作所

发明人： 中田　百科

IPC分类号： G06N99/00 ,  G06F17/10 ,  G06N20/10

摘要： 機械学習で導出する非線形性の強い目的関数をイジングモデルへ変換することで、アリーリング等による最適解の探索を可能とする情報処理システムを提供する。本発明は、学習データベースに対して機械学習により目的関数を導出する目的関数導出システムと、前記目的関数を変換する関数変換システムと、を備え、前記目的関数導出システムは、機械学習の手法を設定する機械学習設定部と、前記目的関数を導出する学習部と、を有し、前記関数変換システムは、ダミー変数の生成方法を設定するダミー変数設定部と、前記ダミー変数を生成するダミー変数生成部と、前記目的関数に陽に現れる説明変数を前記ダミー変数を用いて消去することで、前記説明変数の二次より高次の非線形項を二次以下に次元を落とし、前記目的関数を前記ダミー変数および目的変数に関する前記制約なし二次形式関数または前記線形制約あり一次形式関数へ変換する関数変換部と、を有する。