公开(公告)号：US20220391763A1

公开(公告)日：2022-12-08

申请号：US17811555

申请日：2022-07-08

Applicant: Amazon Technologies, Inc.

Inventor： Leo Parker Dirac ,  Nicolle M. Correa ,  Aleksandr Mikhaylovich Ingerman ,  Sriram Krishnan ,  Jin Li ,  Sudhakar Rao Puvvadi ,  Saman Zarandioon

IPC: G06N20/00

Abstract: A machine learning service implements programmatic interfaces for a variety of operations on several entity types, such as data sources, statistics, feature processing recipes, models, and aliases. A first request to perform an operation on an instance of a particular entity type is received, and a first job corresponding to the requested operation is inserted in a job queue. Prior to the completion of the first job, a second request to perform another operation is received, where the second operation depends on a result of the operation represented by the first job. A second job, indicating a dependency on the first job, is stored in the job queue. The second job is initiated when the first job completes.

公开(公告)号：US20220335338A1

公开(公告)日：2022-10-20

申请号：US17810554

申请日：2022-07-01

Applicant: Amazon Technologies, Inc.

Inventor： Leo Parker Dirac ,  Nicolle M. Correa ,  Charles Eric Dannaker

IPC: G06N20/00

Abstract: At a machine learning service, a set of candidate variables that can be used to train a model is identified, including at least one processed variable produced by a feature processing transformation. A cost estimate indicative of an effect of implementing the feature processing transformation on a performance metric associated with a prediction goal of the model is determined. Based at least in part on the cost estimate, a feature processing proposal that excludes the feature processing transformation is implemented.

公开(公告)号：US20190050756A1

公开(公告)日：2019-02-14

申请号：US16159441

申请日：2018-10-12

Applicant: Amazon Technologies, Inc.

Inventor： Leo Parker Dirac ,  Nicolle M. Correa ,  Aleksandr Mikhaylovich Ingerman ,  Sriram Krishnan ,  Jin Li ,  Sudhakar Rao Puvvadi ,  Saman Zarandioon

IPC: G06N99/00

Abstract: A machine learning service implements programmatic interfaces for a variety of operations on several entity types, such as data sources, statistics, feature processing recipes, models, and aliases. A first request to perform an operation on an instance of a particular entity type is received, and a first job corresponding to the requested operation is inserted in a job queue. Prior to the completion of the first job, a second request to perform another operation is received, where the second operation depends on a result of the operation represented by the first job. A second job, indicating a dependency on the first job, is stored in the job queue. The second job is initiated when the first job completes.

公开(公告)号：US10169715B2

公开(公告)日：2019-01-01

申请号：US14489448

申请日：2014-09-17

Applicant: Amazon Technologies, Inc.

Inventor： Leo Parker Dirac ,  Nicolle M. Correa ,  Charles Eric Dannaker

IPC: G06N99/00 ,  H04L29/08

Abstract: At a machine learning service, a set of candidate variables that can be used to train a model is identified, including at least one processed variable produced by a feature processing transformation. A cost estimate indicative of an effect of implementing the feature processing transformation on a performance metric associated with a prediction goal of the model is determined. Based at least in part on the cost estimate, a feature processing proposal that excludes the feature processing transformation is implemented.