-
Patent Agency Ranking
公开(公告)号：US12020133B2
公开(公告)日：2024-06-25
申请号：US18082476
申请日：2022-12-15
Applicant: Apple Inc.
Inventor： Moises Goldszmidt , Anatoly D. Adamov , Juan C. Garcia , Julia R. Reisler , Timothy S. Paek , Vishwas Kulkarni , Yu-Chung Hsiao , Pavan Chitta
IPC: G06N20/20 , G06F18/21 , G06F18/214 , G06F18/2415 , G06N20/00
CPC classification number: G06N20/20 , G06F18/214 , G06F18/217 , G06F18/2415 , G06N20/00
Abstract: Systems and methods are disclosed for triggering an update to a machine-learning model upon detecting that a distribution of particular (e.g., recently collected) input data set is sufficiently different from a distribution training input data set used to train the model. The distributions may be determined to be sufficiently different when a classifier can identify to which distribution individual data elements belong (e.g., to at least a predetermined degree). An update to the machine-learning model can include morphing weights used by the model and/or retraining the model.
-
公开(公告)号：US20230124380A1
公开(公告)日：2023-04-20
申请号：US18082476
申请日：2022-12-15
Applicant: Apple Inc.
Inventor： Moises Goldszmidt , Anatoly D. Adamov , Juan C. Garcia , Julia R. Reisler , Timothy S. Paek , Vishwas Kulkarni , Yu-Chung Hsiao , Pavan Chitta
IPC: G06N20/00 , G06F18/214 , G06F18/21 , G06F18/2415
Abstract: Systems and methods are disclosed for triggering an update to a machine-learning model upon detecting that a distribution of particular (e.g., recently collected) input data set is sufficiently different from a distribution training input data set used to train the model. The distributions may be determined to be sufficiently different when a classifier can identify to which distribution individual data elements belong (e.g., to at least a predetermined degree). An update to the machine-learning model can include morphing weights used by the model and/or retraining the model.
-
公开(公告)号：US20210224687A1
公开(公告)日：2021-07-22
申请号：US16875825
申请日：2020-05-15
Applicant: Apple Inc.
Inventor： Moises Goldszmidt , Anatoly D. Adamov , Juan C. Garcia , Julia R. Reisler , Timothy S. Paek , Vishwas Kulkarni , Yu-Chung Hsiao , Pavan Chitta
Abstract: Systems and methods are disclosed for triggering an update to a machine-learning model upon detecting that a distribution of particular (e.g., recently collected) input data set is sufficiently different from a distribution training input data set used to train the model. The distributions may be determined to be sufficiently different when a classifier can identify to which distribution individual data elements belong (e.g., to at least a predetermined degree). An update to the machine-learning model can include morphing weights used by the model and/or retraining the model.
-
公开(公告)号：US20240338612A1
公开(公告)日：2024-10-10
申请号：US18745654
申请日：2024-06-17
Applicant: Apple Inc.
Inventor： Moises Goldszmidt , Anatoly D. Adamov , Juan C. Garcia , Julia R. Reisler , Timothy S. Paek , Vishwas Kulkarni , Yu-Chung Hsiao , Pavan Chitta
IPC: G06N20/20 , G06F18/21 , G06F18/214 , G06F18/2415 , G06N20/00
CPC classification number: G06N20/20 , G06F18/214 , G06F18/217 , G06F18/2415 , G06N20/00
Abstract: Systems and methods are disclosed for triggering an update to a machine-learning model upon detecting that a distribution of particular (e.g., recently collected) input data set is sufficiently different from a distribution training input data set used to train the model. The distributions may be determined to be sufficiently different when a classifier can identify to which distribution individual data elements belong (e.g., to at least a predetermined degree). An update to the machine-learning model can include morphing weights used by the model and/or retraining the model.
-
公开(公告)号：US11562297B2
公开(公告)日：2023-01-24
申请号：US16875825
申请日：2020-05-15
Applicant: Apple Inc.
Inventor： Moises Goldszmidt , Anatoly D. Adamov , Juan C. Garcia , Julia R. Reisler , Timothy S. Paek , Vishwas Kulkarni , Yu-Chung Hsiao , Pavan Chitta
Abstract: Systems and methods are disclosed for triggering an update to a machine-learning model upon detecting that a distribution of particular (e.g., recently collected) input data set is sufficiently different from a distribution training input data set used to train the model. The distributions may be determined to be sufficiently different when a classifier can identify to which distribution individual data elements belong (e.g., to at least a predetermined degree). An update to the machine-learning model can include morphing weights used by the model and/or retraining the model.
-
-
-
-
Search Results
5
records
(took 0.015 seconds)
