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

WO2022097131A1 A SYSTEM AND METHOD FOR THE UNIFICATION AND OPTIMIZATION OF MACHINE LEARNING INFERENCE PIPELINES 审中-公开
公开(公告)号：WO2022097131A1
公开(公告)日：2022-05-12
申请号：PCT/IL2021/051188
申请日：2021-10-03
申请人： SAFERIDE TECHNOLOGIES LTD.
发明人： STEIN, Yehiel , VARDI, Yossi , APARTSIN, Alexander
IPC分类号： G06N5/04
摘要： A system comprising a processing circuitry configured to: obtain one or more MLIPs, each comprised of a sequence of one or more Data Processing Elements (DPEs), and each having (a) at least one input provided to the respective DPE, and (b) at least one output provided by the respective DPE, wherein the output of a given DPE of the DPEs, is the input of a subsequent DPE of the sequence, and wherein at least one of the DPEs is a trained machine learning model; generate, for each of the MLIPs, a respective pipeline representation comprising representations of the sequence, based on the DPEs, the inputs of the DPEs, and the outputs of the DPEs; merge the plurality of MLIP representations into a common representation; optimize the common representation; and generate, based on the common representation, a target model consuming less resources than the MLIPs.

2. 发明公开

EP4241210A1 A SYSTEM AND METHOD FOR THE UNIFICATION AND OPTIMIZATION OF MACHINE LEARNING INFERENCE PIPELINES 审中-公开
公开(公告)号：EP4241210A1
公开(公告)日：2023-09-13
申请号：EP21888823.8
申请日：2021-10-03
申请人： Saferide Technologies Ltd.
发明人： STEIN, Yehiel , VARDI, Yossi , APARTSIN, Alexander
IPC分类号： G06N5/04

3. 发明公开

EP4270251A1 A SYSTEM AND METHOD FOR MODEL TRAINING FOR A NEW SIGNAL BASED ON METADATA REPRESENTATIONS OF KNOWN SIGNALS 审中-公开
公开(公告)号：EP4270251A1
公开(公告)日：2023-11-01
申请号：EP23168894.6
申请日：2023-04-20
申请人： Saferide Technologies Ltd.
发明人： HENRICHS, Kevin , BLANC, Sebastian Michael , SHALEV, Yuval , APARTSIN, Alexander
IPC分类号： G06N3/0455 , G06N3/09 , G06N5/02
摘要： A system comprising a processing circuitry configured to: obtain: (a) one or more metadata representations, each representing metadata relating to a signal type of one or more signal types; (b) one or more signal sequences, each signal sequence is an ordered sequence of values associated with a given signal type of the one or more signal types; and (c) at least one new signal sequence, being a new ordered sequence of values, each associated with a label, the at least one new signal sequence is associated with corresponding at least one new signal type, not included in the one or more signal types; train a meta learner autoencoder, capable of mapping at least one given signal sequence and at least one respective metadata representation, being the metadata representation representing the signal type of the given signal sequence into a meta representation vector, wherein the trained meta learner autoencoder comprises a meta learner encoder and a meta learner decoder; determine, based on the metadata representations, a predicted metadata representation representing the at least one new signal type; and train, by utilizing the meta representation vector mapped by the meta learner encoder from the at least one new signal sequence and the predicted metadata representation, a new task model, capable of receiving one or more unlabeled signal sequences associated with the at least one new signal type, and predicting, utilizing the predicted metadata representation and the meta learner encoder, for each of the unlabeled signal sequences, a corresponding label.

4. 发明公开

EP4109354A1 A SYSTEM AND METHOD FOR PHYSICAL MODEL BASED MACHINE LEARNING 审中-公开
公开(公告)号：EP4109354A1
公开(公告)日：2022-12-28
申请号：EP22175876.6
申请日：2022-05-27
申请人： Saferide Technologies Ltd.
发明人： APARTSIN, Alexander , STEIN, Yehiel , VARDI, Yossi
IPC分类号： G06N20/00 , G06F30/27
摘要： A physics-based model machine learning system, the physics-based model machine learning system comprising a processing circuitry (230) configured to: obtain: (a) a training data-set, the training data-set comprising a plurality of training records, each training record including a collection of features describing a given allowed state of a physical entity (310), and (b) one or more physical models, modeling allowed physical patterns associated with the physical entity; enrich the training data-set by determining values of one or more unobservable features for one or more given training records of the training records, wherein the unobservable features are determined utilizing at least one of the physical models and at least one of the features of the respective given training records, giving rise to an enriched training data-set (320); train, using the enriched training data-set (330), a machine learning model capable of receiving one or more inference records, and determining, for each of the inference records, a corresponding normality score being indicative of conformity of the respective inference record with an allowed state of the physical entity; and classify, using the machine learning model, an incoming record describing a state of the physical entity at a given time, as abnormal upon the normality score determined by the machine learning model being below a threshold (340).

5. 发明公开

EP4235515A1 A SYSTEM AND METHOD FOR MODEL CONFIGURATION SELECTION 审中-公开
公开(公告)号：EP4235515A1
公开(公告)日：2023-08-30
申请号：EP23159156.1
申请日：2023-02-28
申请人： Saferide Technologies Ltd.
发明人： KOVALETS, Sofiia , BARABANOV, Stanislav , SHALEV, Yuval , APARTSIN, Alexander
IPC分类号： G06N3/0985 , G06N3/0895 , G06N3/10 , G06N3/04 , G06N20/00
摘要： A model configuration selection system comprising a processing circuitry configured to:
(A) obtain: (a) one or more model configurations and (b) a training data-set comprising a plurality of unlabeled records;
(B) cluster the training data-set into two or more training data-set clusters using a clustering algorithm;
(C) label (a) the unlabeled records of a subset of the training data-set clusters with a synthetic normal label, and (b) the unlabeled records of the training data-set clusters not included in the subset with a synthetic abnormal label;
(D) train, for each model configuration, a corresponding model utilizing the corresponding set of parameters;
(E) determine, for each model, a score, associated with an ability of the corresponding model to determine labels to the unlabeled records of the training data-set in accordance with the synthetic normal labels and with the synthetic abnormal labels; and
(F) perform an action, based on the scores.

6. 发明公开

EP4099623A1 A SYSTEM AND METHOD FOR FEATURE SELECTION RECOMMENDATION 审中-公开
公开(公告)号：EP4099623A1
公开(公告)日：2022-12-07
申请号：EP22175859.2
申请日：2022-05-27
申请人： Saferide Technologies Ltd.
发明人： BAKALO, Ran , APARTSIN, Alexander , STEIN, Yehiel , VARDI, Yossi
IPC分类号： H04L9/40
摘要： A feature selection recommendation system, the feature selection recommendation system comprising a processing circuitry configured to: obtain: (a) a training data-set, the training data-set comprising a plurality of records, each record including a collection of features describing a given allowed state of a physical entity, and (b) a selection of one or more selected features of the features; generate, using a causality discovery model, for a plurality of pairs of the features of the training data-set, a respective causality score, the causality score being indicative of an influence between the features of the respective pair; identify additional recommended features, being one or more features that comply with a recommendation condition based on the plurality of pairs and the causality scores generated for the pairs; and provide a user of the feature selection recommendation system with an indication of the additional recommended features.

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