-
公开(公告)号：US10387749B2
公开(公告)日：2019-08-20
申请号：US15710377
申请日：2017-09-20
Applicant: Google Inc.
Inventor： Yair Movshovitz-Attias , King Hong Leung , Saurabh Singh , Alexander Toshev , Sergey Ioffe
Abstract: The present disclosure provides systems and methods that enable distance metric learning using proxies. A machine-learned distance model can be trained in a proxy space in which a loss function compares an embedding provided for an anchor data point of a training dataset to a positive proxy and one or more negative proxies, where each of the positive proxy and the one or more negative proxies serve as a proxy for two or more data points included in the training dataset. Thus, each proxy can approximate a number of data points, enabling faster convergence. According to another aspect, the proxies of the proxy space can themselves be learned parameters, such that the proxies and the model are trained jointly. Thus, the present disclosure enables faster convergence (e.g., reduced training time). The present disclosure provides example experiments which demonstrate a new state of the art on several popular training datasets.
-
公开(公告)号：US20190065957A1
公开(公告)日：2019-02-28
申请号：US15690426
申请日：2017-08-30
Applicant: Google Inc.
Inventor： Yair Movshovitz-Attias , King Hong Leung , Saurabh Singh , Alexander Toshev , Sergey Ioffe
Abstract: The present disclosure provides systems and methods that enable distance metric learning using proxies. A machine-learned distance model can be trained in a proxy space in which a loss function compares an embedding provided for an anchor data point of a training dataset to a positive proxy and one or more negative proxies, where each of the positive proxy and the one or more negative proxies serve as a proxy for two or more data points included in the training dataset. Thus, each proxy can approximate a number of data points, enabling faster convergence. According to another aspect, the proxies of the proxy space can themselves be learned parameters, such that the proxies and the model are trained jointly. Thus, the present disclosure enables faster convergence (e.g., reduced training time). The present disclosure provides example experiments which demonstrate a new state of the art on several popular training datasets.
-
公开(公告)号：US20190065899A1
公开(公告)日：2019-02-28
申请号：US15710377
申请日：2017-09-20
Applicant: Google Inc.
Inventor： Yair Movshovitz-Attias , King Hong Leung , Saurabh Singh , Alexander Toshev , Sergey Ioffe
CPC classification number: G06K9/6215 , G06K9/4628 , G06K9/6232 , G06K9/6255 , G06K9/6256 , G06K9/6262 , G06K9/66 , G06N20/00
Abstract: The present disclosure provides systems and methods that enable distance metric learning using proxies. A machine-learned distance model can be trained in a proxy space in which a loss function compares an embedding provided for an anchor data point of a training dataset to a positive proxy and one or more negative proxies, where each of the positive proxy and the one or more negative proxies serve as a proxy for two or more data points included in the training dataset. Thus, each proxy can approximate a number of data points, enabling faster convergence. According to another aspect, the proxies of the proxy space can themselves be learned parameters, such that the proxies and the model are trained jointly. Thus, the present disclosure enables faster convergence (e.g., reduced training time). The present disclosure provides example experiments which demonstrate a new state of the art on several popular training datasets.
-
-
Search Results
3
records
(took 0.013 seconds)
