知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

10 records (took 0.017 seconds)

    Systems, methods, and apparatuses for implementing machine learning model training and deployment with a rollback mechanism
    2.
    发明授权
    Systems, methods, and apparatuses for implementing machine learning model training and deployment with a rollback mechanism 有权

    公开(公告)号：US10713594B2

    公开(公告)日：2020-07-14

    申请号：US15407147

    申请日：2017-01-16

    Applicant: salesforce.com, inc.

    Inventor： Kit Pang Szeto ,  Simon Chan

    IPC: G06N20/00 ,  G06F11/14 ,  G06Q30/02

    Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing machine learning model training and deployment with a rollback mechanism within a computing environment. For example, an exemplary machine learning platform includes means for receiving training data as input at the machine learning platform, in which the training data includes a multiple transactions, each of the transactions specifying a plurality of features upon which to make a prediction and a label representing a correct answer for the plurality of features according to each respective transaction; specifying a model to be trained by the machine learning platform using the training data, in which the model includes a plurality of algorithms and source code; generating a new predictive engine variant by training the model to algorithmically arrive upon the label representing the correct answer as provided with the training data based on the plurality of features for each of the multiple transactions; versioning the new predictive engine variant based at least on the time the new predictive engine variant was generated a version of the source code utilized within the model and the training data received as input; deploying the new predictive engine variant into a production environment to replace a prior version of the predictive engine variant; and rolling back the new predictive engine variant from the production environment to a specified version which is less than a version of the new predictive engine variant. Other related embodiments are disclosed.

    Methods and systems for predictive engine evaluation and replay of engine performance
    4.
    发明授权
    Methods and systems for predictive engine evaluation and replay of engine performance 有权

    公开(公告)号：US10192172B2

    公开(公告)日：2019-01-29

    申请号：US15404052

    申请日：2017-01-11

    Applicant: salesforce.com, inc.

    Inventor： Ka Hou Chan ,  Simon Chan ,  Kit Pang Szeto ,  Yue Kwen Justin Yip

    IPC: G06N99/00 ,  G06Q30/02 ,  G06F17/30 ,  G06N7/00 ,  G06N5/02

    Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing predictive engine evaluation and replay of engine performance. An exemplary system may include, for example: means selecting a first set of one or more algorithms for a machine learning model; tuning a first group of predictive engine parameters for the machine learning model; training the machine learning model with one or more sources of data using the selected first set of one or more algorithms and the first group of tuned predictive engine parameters to generate a first predictive engine variant from the trained machine learning model; selecting a second set of one or more algorithms for a machine learning model which are different than the first set; tuning a second group of predictive engine parameters for the machine learning model which are different than the first group; training the machine learning model with the one or more sources of data using the selected second set of one or more algorithms and the second group of tuned predictive engine parameters to generate a second predictive engine variant from the trained machine learning model; performing multiple experiments using the first and second predictive engine variants; comparing results from the multiple experiments; and deploying either the first predictive engine variant or the second predictive engine variant based on the comparison of the results of the multiple experiments. Other related embodiments are disclosed.

    FRAMEWORK FOR MANAGEMENT OF MODELS BASED ON TENANT BUSINESS CRITERIA IN AN ON-DEMAND ENVIRONMENT
    6.
    发明申请
    FRAMEWORK FOR MANAGEMENT OF MODELS BASED ON TENANT BUSINESS CRITERIA IN AN ON-DEMAND ENVIRONMENT 审中-公开

    公开(公告)号：US20180096028A1

    公开(公告)日：2018-04-05

    申请号：US15713069

    申请日：2017-09-22

    Applicant: salesforce.com, inc.

    Inventor： Chalenge Masekera ,  Simon Chan ,  Kit Pang Szeto

    IPC: G06F17/30 ,  G06F21/62

    CPC classification number: G06F16/2453 ,  G06F16/2462 ,  G06F21/6227

    Abstract: In accordance with embodiments, there are provided mechanisms and methods for facilitating a framework for management of machine learning models for tenants in an on-demand services environment according to one embodiment. In one embodiment and by way of example, a method comprises determining, by a model management server computing device (“management device”), business criteria for a tenant in a multi-tenant environment, where the business criteria are based on business preferences of the tenant. The method may further include building, by the management device, multiple models dedicated to the tenant based on the business criteria such that each model is trained and fitted to perform one or more combinations of processes based on one or more integrations of the business criteria. The method may further include dynamically selecting, by the management device, a model from the multiple models to perform a combination of processes involving an integration of two or more criterion of the business criteria as requested by the tenant.

    METHODS AND SYSTEMS FOR PREDICTIVE ENGINE EVALUATION AND REPLAY OF ENGINE PERFORMANCE
    7.
    发明申请
    METHODS AND SYSTEMS FOR PREDICTIVE ENGINE EVALUATION AND REPLAY OF ENGINE PERFORMANCE 审中-公开
    用于预测发动机评估和发动机性能替代的方法和系统

    公开(公告)号：US20160275406A1

    公开(公告)日：2016-09-22

    申请号：US14997662

    申请日：2016-01-18

    Applicant: salesforce.com, inc.

    Inventor： Ka Hou Chan ,  Simon Chan ,  Kit Pang Szeto ,  Yue Kwen Justin Yip

    IPC: G06N7/00 ,  G06F17/30

    CPC classification number: G06N20/00 ,  G06F16/2455 ,  G06F16/248 ,  G06N5/022 ,  G06N7/005 ,  G06Q30/0204

    Abstract: Disclosed are methods and systems of tracking the deployment of a predictive engine for machine learning, including steps to deploy an engine variant of the predictive engine based on an engine parameter set, wherein the engine parameter set identifies at least one data source and at least one algorithm; receive one or more queries to the deployed engine variant from one or more end-user devices, and in response, generate predicted results; receive one or more actual results corresponding to the predicted results; associate the queries, the predicted results, and the actual results with a replay tag, and record them with the corresponding deployed engine variant.

    Abstract translation: 公开了跟踪用于机器学习的预测引擎的部署的方法和系统，包括基于引擎参数集部署预测引擎的引擎变体的步骤，其中引擎参数集识别至少一个数据源和至少一个 算法; 从一个或多个最终用户设备接收对所部署的引擎变体的一个或多个查询，作为响应，生成预测结果; 接收与预测结果相对应的一个或多个实际结果; 使用重播标签关联查询，预测结果和实际结果，并使用相应的部署引擎变体进行记录。

    SYSTEMS, METHODS, AND APPARATUSES FOR IMPLEMENTING MACHINE LEARNING MODEL TRAINING AND DEPLOYMENT WITH A ROLLBACK MECHANISM
    9.
    发明申请
    SYSTEMS, METHODS, AND APPARATUSES FOR IMPLEMENTING MACHINE LEARNING MODEL TRAINING AND DEPLOYMENT WITH A ROLLBACK MECHANISM 审中-公开

    公开(公告)号：US20170124487A1

    公开(公告)日：2017-05-04

    申请号：US15407147

    申请日：2017-01-16

    Applicant: salesforce.com, inc.

    Inventor： Kit Pang Szeto ,  Simon Chan

    IPC: G06N99/00 ,  G06F11/14 ,  G06N5/04

    CPC classification number: G06N20/00 ,  G06F11/1448 ,  G06Q30/0204

    Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing machine learning model training and deployment with a rollback mechanism within a computing environment. For example, an exemplary machine learning platform includes means for receiving training data as input at the machine learning platform, in which the training data includes a multiple transactions, each of the transactions specifying a plurality of features upon which to make a prediction and a label representing a correct answer for the plurality of features according to each respective transaction; specifying a model to be trained by the machine learning platform using the training data, in which the model includes a plurality of algorithms and source code; generating a new predictive engine variant by training the model to algorithmically arrive upon the label representing the correct answer as provided with the training data based on the plurality of features for each of the multiple transactions; versioning the new predictive engine variant based at least on the time the new predictive engine variant was generated a version of the source code utilized within the model and the training data received as input; deploying the new predictive engine variant into a production environment to replace a prior version of the predictive engine variant; and rolling back the new predictive engine variant from the production environment to a specified version which is less than a version of the new predictive engine variant. Other related embodiments are disclosed.

    METHODS AND SYSTEMS FOR PREDICTIVE ENGINE EVALUATION AND REPLAY OF ENGINE PERFORMANCE
    10.
    发明申请
    METHODS AND SYSTEMS FOR PREDICTIVE ENGINE EVALUATION AND REPLAY OF ENGINE PERFORMANCE 审中-公开

    公开(公告)号：US20170124486A1

    公开(公告)日：2017-05-04

    申请号：US15404052

    申请日：2017-01-11

    Applicant: salesforce.com, inc.

    Inventor： Ka Hou Chan ,  Simon Chan ,  Kit Pang Szeto ,  Yue Kwen Justin Yip

    IPC: G06N99/00 ,  G06N5/02

    CPC classification number: G06N99/005 ,  G06F17/30477 ,  G06F17/30554 ,  G06N5/022 ,  G06N7/005 ,  G06Q30/0204

    Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing predictive engine evaluation and replay of engine performance. An exemplary system may include, for example: means selecting a first set of one or more algorithms for a machine learning model; tuning a first group of predictive engine parameters for the machine learning model; training the machine learning model with one or more sources of data using the selected first set of one or more algorithms and the first group of tuned predictive engine parameters to generate a first predictive engine variant from the trained machine learning model; selecting a second set of one or more algorithms for a machine learning model which are different than the first set; tuning a second group of predictive engine parameters for the machine learning model which are different than the first group; training the machine learning model with the one or more sources of data using the selected second set of one or more algorithms and the second group of tuned predictive engine parameters to generate a second predictive engine variant from the trained machine learning model; performing multiple experiments using the first and second predictive engine variants; comparing results from the multiple experiments; and deploying either the first predictive engine variant or the second predictive engine variant based on the comparison of the results of the multiple experiments. Other related embodiments are disclosed.

Patent Agency Ranking