Graph-based transfer learning
    3.
    发明授权
    Graph-based transfer learning 有权
    基于图形的传输学习

    公开(公告)号:US09477929B2

    公开(公告)日:2016-10-25

    申请号:US13619142

    申请日:2012-09-14

    IPC分类号: G06F5/00 G06N5/00 G06N99/00

    CPC分类号: G06N99/005

    摘要: Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. A graph-based transfer learning framework propagates label information from a source domain to a target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bipartite graph. An iterative algorithm renders the framework scalable to large-scale applications. The framework propagates the label information to both features irrelevant to the source domain and unlabeled examples in the target domain via common features in a principled way.

    摘要翻译: 转移学习是利用来自某些领域的标记示例的信息来预测另一个域中的示例的标签的任务。 发现情绪预测,图像分类和网络入侵检测等丰富的实际应用。 基于图形的传输学习框架通过示例特征示例三方图将标签信息从源域传播到目标域,并通过示例性的二分图更加强调来自目标域的标记示例。 迭代算法使框架可扩展到大规模应用程序。 该框架通过原理方式的共同特征将标签信息传播到与源域无关的特征和目标域中的未标记示例。

    Graph-based framework for multi-task multi-view learning
    4.
    发明授权
    Graph-based framework for multi-task multi-view learning 有权
    基于图形的多任务多视图学习框架

    公开(公告)号:US08990128B2

    公开(公告)日:2015-03-24

    申请号:US13488885

    申请日:2012-06-05

    IPC分类号: G06F15/18 G06K9/62

    CPC分类号: G06K9/628

    摘要: A system and method a Multi-Task Multi-View (M2TV) learning problem. The method uses the label information from related tasks to make up for the lack of labeled data in a single task. The method further uses the consistency among different views to improve the performance. It is tailored for the above complicated dual heterogeneous problems where multiple related tasks have both shared and task-specific views (features), since it makes full use of the available information.

    摘要翻译: 多任务多视图(M2TV)学习问题的系统和方法。 该方法使用相关任务的标签信息来弥补单个任务中缺少标记数据。 该方法进一步使用不同视图之间的一致性来提高性能。 它针对上述复杂的双重异构问题,其中多个相关任务具有共享和任务特定的视图(特征),因为它充分利用了可用的信息。

    GRAPH-BASED FRAMEWORK FOR MULTI-TASK MULTI-VIEW LEARNING
    5.
    发明申请
    GRAPH-BASED FRAMEWORK FOR MULTI-TASK MULTI-VIEW LEARNING 有权
    用于多任务多视图学习的基于图形的框架

    公开(公告)号:US20130325756A1

    公开(公告)日:2013-12-05

    申请号:US13488885

    申请日:2012-06-05

    IPC分类号: G06F15/18

    CPC分类号: G06K9/628

    摘要: A system and method a Multi-Task Multi-View (M2TV) learning problem. The method uses the label information from related tasks to make up for the lack of labeled data in a single task. The method further uses the consistency among different views to improve the performance. It is tailored for the above complicated dual heterogeneous problems where multiple related tasks have both shared and task-specific views (features), since it makes full use of the available information.

    摘要翻译: 多任务多视图(M2TV)学习问题的系统和方法。 该方法使用相关任务的标签信息来弥补单个任务中缺少标记数据。 该方法进一步使用不同视图之间的一致性来提高性能。 它针对上述复杂的双重异构问题,其中多个相关任务具有共享和任务特定的视图(特征),因为它充分利用了可用的信息。

    GRAPH-BASED TRANSFER LEARNING
    6.
    发明申请
    GRAPH-BASED TRANSFER LEARNING 审中-公开
    基于图形的传输学习

    公开(公告)号:US20130013540A1

    公开(公告)日:2013-01-10

    申请号:US13619142

    申请日:2012-09-14

    IPC分类号: G06F15/18

    CPC分类号: G06N99/005

    摘要: Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. A graph-based transfer learning framework propagates label information from a source domain to a target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bipartite graph. An iterative algorithm renders the framework scalable to large-scale applications. The framework propagates the label information to both features irrelevant to the source domain and unlabeled examples in the target domain via common features in a principled way.

    摘要翻译: 转移学习是利用来自某些领域的标记示例的信息来预测另一个域中的示例的标签的任务。 发现情绪预测,图像分类和网络入侵检测等丰富的实际应用。 基于图形的传输学习框架通过示例特征示例三方图将标签信息从源域传播到目标域,并通过示例性的二分图更加强调来自目标域的标记示例。 迭代算法使框架可扩展到大规模应用程序。 该框架通过原理方式的共同特征将标签信息传播到与源域无关的特征和目标域中的未标记示例。

    Graph-based transfer learning
    7.
    发明申请
    Graph-based transfer learning 审中-公开
    基于图形的传输学习

    公开(公告)号:US20110320387A1

    公开(公告)日:2011-12-29

    申请号:US12938063

    申请日:2010-11-02

    IPC分类号: G06F15/18

    CPC分类号: G06N20/00

    摘要: Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. A graph-based transfer learning framework propagates label information from a source domain to a target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bipartite graph. An iterative algorithm renders the framework scalable to large-scale applications. The framework propagates the label information to both features irrelevant to the source domain and unlabeled examples in the target domain via common features in a principled way.

    摘要翻译: 转移学习是利用来自某些领域的标记示例的信息来预测另一个域中的示例的标签的任务。 发现情绪预测,图像分类和网络入侵检测等丰富的实际应用。 基于图形的传输学习框架通过示例特征示例三方图将标签信息从源域传播到目标域,并通过示例性的二分图更加强调来自目标域的标记示例。 迭代算法使框架可扩展到大规模应用程序。 该框架通过原理方式的共同特征将标签信息传播到与源域无关的特征和目标域中的未标记示例。

    Method and apparatus for hierarchical wafer quality predictive modeling
    9.
    发明授权
    Method and apparatus for hierarchical wafer quality predictive modeling 有权
    分层晶圆质量预测模型的方法和装置

    公开(公告)号:US08732627B2

    公开(公告)日:2014-05-20

    申请号:US13526152

    申请日:2012-06-18

    IPC分类号: G06F17/50

    摘要: A method for performing enhanced wafer quality prediction in a semiconductor manufacturing process includes the steps of: obtaining data including at least one of tensor format wafer processing conditions, historical wafer quality measurements and prior knowledge relating to at least one of the semiconductor manufacturing process and wafer quality; building a hierarchical prediction model including at least the tensor format wafer processing conditions; and predicting wafer quality for a newly fabricated wafer based at least on the hierarchical prediction model and corresponding tensor format wafer processing conditions.

    摘要翻译: 一种用于在半导体制造过程中执行增强的晶片质量预测的方法包括以下步骤:获得包括张量格式晶片处理条件,历史晶片质量测量和与半导体制造工艺和晶片中的至少一个相关的先前知识中的至少一个的数据 质量; 构建包括至少张量格式晶片处理条件的分级预测模型; 并且至少基于分层预测模型和对应的张量格式晶片处理条件来预测新制造的晶片的晶片质量。