Accuracy of Classification Models

    公开(公告)号:US20210174148A1

    公开(公告)日:2021-06-10

    申请号:US17107851

    申请日:2020-11-30

    Applicant: Orbotech Ltd.

    Abstract: An Artificial Intelligence system, an apparatus and, a computer program product and a method for automatic improvement of artificial intelligence classification models. A model-performance measurement of the classification model is iteratively improved by at least a predetermined target goal in each iteration. The iterative improvement comprises generating a hypotheses graph for improving the classification model, based on a list of hypotheses and scores thereof. Each hypothesis relates to a strategy for potentially improving the classification model, and is associated with a score indicating a likelihood that an application thereof improves the model-performance measurement. Each node of the hypotheses graph comprises a hypothesis of the list of hypotheses. The iterative improvement further comprises selecting a selected hypothesis from the hypotheses graph based on a traversal thereof; and executing the selected hypothesis thereby updating the classification model and improving the model-performance measurement by at least the predetermined target goal.

    Adaptive Routing for Correcting Die Placement Errors

    公开(公告)号:US20220012405A1

    公开(公告)日:2022-01-13

    申请号:US17288369

    申请日:2019-10-02

    Applicant: Orbotech Ltd.

    Abstract: A method includes, receiving a layout design of at least part of an electronic module, the design specifying at least (i) an electronic device coupled to at least a substrate, and (ii) an electrical trace that is connected to the electronic device and has a designed route. A digital input, which represents at least part of an actual electronic module that was manufactured in accordance with the layout design but without at least a portion of the electrical trace, is received. An error in coupling the electronic device to the substrate, relative to the layout design, is estimated based on the digital input. An actual route that corrects the estimated error, is calculated for at least the portion of the electrical trace. At least the portion of the electrical trace is formed on the substrate of the actual electronic module, along the actual route instead of the designed route.

    ADAPTIVE LEARNING FOR IMAGE CLASSIFICATION

    公开(公告)号:US20210295213A1

    公开(公告)日:2021-09-23

    申请号:US17164756

    申请日:2021-02-01

    Applicant: Orbotech Ltd.

    Inventor: Gonen Raveh

    Abstract: A method, a computerized apparatus and a computer program product for adaptive learning for image classification. The method comprises applying a set of classification models on a calibration dataset and on a production dataset, and calculating disagreement measurements over the predictions thereof on each dataset. Based on similarity measurement between the disagreement measurement of the calibration and the production datasets, being below a predetermined threshold, a data drift is indicated in the production dataset. The method further comprises determining a training dataset for training a classification model for the production dataset. The training dataset is selected over a plurality of sets of images ordered according to time intervals in which images therein are obtained. The selection is performed based on weights determined for the plurality of sets.

    Adaptive routing for correcting die placement errors

    公开(公告)号:US11652008B2

    公开(公告)日:2023-05-16

    申请号:US17288369

    申请日:2019-10-02

    Applicant: Orbotech Ltd.

    CPC classification number: G06F30/3953 G06F30/31 G06F30/398 G06F2115/12

    Abstract: A method includes, receiving a layout design of at least part of an electronic module, the design specifying at least (i) an electronic device coupled to at least a substrate, and (ii) an electrical trace that is connected to the electronic device and has a designed route. A digital input, which represents at least part of an actual electronic module that was manufactured in accordance with the layout design but without at least a portion of the electrical trace, is received. An error in coupling the electronic device to the substrate, relative to the layout design, is estimated based on the digital input. An actual route that corrects the estimated error, is calculated for at least the portion of the electrical trace. At least the portion of the electrical trace is formed on the substrate of the actual electronic module, along the actual route instead of the designed route.

    Accuracy of classification models

    公开(公告)号:US11640559B2

    公开(公告)日:2023-05-02

    申请号:US17107851

    申请日:2020-11-30

    Applicant: Orbotech Ltd.

    Abstract: An Artificial Intelligence system, an apparatus and, a computer program product and a method for automatic improvement of artificial intelligence classification models. A model-performance measurement of the classification model is iteratively improved by at least a predetermined target goal in each iteration. The iterative improvement comprises generating a hypotheses graph for improving the classification model, based on a list of hypotheses and scores thereof. Each hypothesis relates to a strategy for potentially improving the classification model, and is associated with a score indicating a likelihood that an application thereof improves the model-performance measurement. Each node of the hypotheses graph comprises a hypothesis of the list of hypotheses. The iterative improvement further comprises selecting a selected hypothesis from the hypotheses graph based on a traversal thereof; and executing the selected hypothesis thereby updating the classification model and improving the model-performance measurement by at least the predetermined target goal.

Patent Agency Ranking