Patents by Inventor Kirill SAVCHENKO

Kirill SAVCHENKO has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Patent number: 12361531
    Abstract: There is provided a method of automated defects' classification, and a system thereof. The method comprises obtaining data informative of a set of defects' physical attributes usable to distinguish between defects of different classes among the plurality of classes; training a first machine learning model to generate, for the given defect, a multi-label output vector informative of values of the physical attributes, thereby generating for the given defect a multi-label descriptor; and using the trained first machine learning model to generate multi-label descriptors of the defects in the specimen. The method can further comprise obtaining data informative of multi-label data sets, each data set being uniquely indicative of a respective class of the plurality of classes and comprising a unique set of values of the physical attributes; and classifying defects in the specimen by matching respectively generated multi-label descriptors of the defects to the multi-label data sets.
    Type: Grant
    Filed: March 24, 2020
    Date of Patent: July 15, 2025
    Assignee: Applied Materials Israel Ltd.
    Inventors: Ohad Shaubi, Boaz Cohen, Kirill Savchenko, Ore Shtalrid
  • Publication number: 20220222806
    Abstract: There is provided a method of automated defects' classification, and a system thereof. The method comprises obtaining data informative of a set of defects' physical attributes usable to distinguish between defects of different classes among the plurality of classes; training a first machine learning model to generate, for the given defect, a multi-label output vector informative of values of the physical attributes, thereby generating for the given defect a multi-label descriptor; and using the trained first machine learning model to generate multi-label descriptors of the defects in the specimen. The method can further comprise obtaining data informative of multi-label data sets, each data set being uniquely indicative of a respective class of the plurality of classes and comprising a unique set of values of the physical attributes; and classifying defects in the specimen by matching respectively generated multi-label descriptors of the defects to the multi-label data sets.
    Type: Application
    Filed: March 24, 2020
    Publication date: July 14, 2022
    Inventors: Ohad SHAUBI, Boaz COHEN, Kirill SAVCHENKO, Ore SHTALRID
  • Patent number: 11151706
    Abstract: A system, method and computer readable medium for classifying defects, the method comprising: receiving classified first defects, and potential defects, each first and potential defect having values for attributes; processing the first and potential defects to select a subset of the attributes that differentiates the first defects from the potential defects; obtaining first and second functions based on the first defects and potential defects, respectively; obtaining a first threshold for the first function, and a second threshold for a combination of the first and second functions; applying the first function and the second function to each potential defect to obtain first and second scores, respectively; and determining a combined score of the first and second scores; and indicating as a defect of a potentially new type a potential defect when the first score is lower than the first threshold or the combined score exceeds the second threshold.
    Type: Grant
    Filed: January 16, 2019
    Date of Patent: October 19, 2021
    Assignee: APPLIED MATERIAL ISRAEL, LTD.
    Inventors: Kirill Savchenko, Assaf Asbag, Boaz Cohen
  • Patent number: 11037286
    Abstract: There are provided a classifier and a method of classifying defects in a semiconductor specimen. The classifier enables assigning each class to a classification group among three or more classification groups with different priorities. Classifier further enables setting purity, accuracy and/or extraction requirements separately for each class, and optimizing the classification results in accordance with per-class requirements. During training, the classifier is configured to generate a classification rule enabling the highest possible contribution of automated classification while meeting per-class quality requirements defined for each class.
    Type: Grant
    Filed: September 28, 2017
    Date of Patent: June 15, 2021
    Assignee: Applied Materials Israel Ltd.
    Inventors: Assaf Asbag, Ohad Shaubi, Kirill Savchenko, Shiran Gan-Or, Boaz Cohen, Zeev Zohar
  • Patent number: 10921334
    Abstract: An examination system, a method of obtaining a training set for a classifier, and a non-transitory computer readable medium, the method comprising: upon receiving in a memory device object inspection results comprising data indicative of potential defects, each potential defect of the potential defects associated with a multiplicity of attribute values defining a location of the potential defect in an attribute space: sampling by the processor a first set of defects from the potential defects, wherein the defects within the first set are dispersed independently of a density of the potential defects in the attribute space; and obtaining by the processor a training defect sample set comprising the first set of defects and data or parameters representative of the density of the potential defects in the attribute space.
    Type: Grant
    Filed: March 22, 2018
    Date of Patent: February 16, 2021
    Assignee: APPLIED MATERIALS ISRAEL LTD.
    Inventors: Kirill Savchenko, Assaf Asbag, Boaz Cohen
  • Publication number: 20200226743
    Abstract: A system, method and computer readable medium for classifying defects, the method comprising: receiving classified first defects, and potential defects, each first and potential defect having values for attributes; processing the first and potential defects to select a subset of the attributes that differentiates the first defects from the potential defects; obtaining first and second functions based on the first defects and potential defects, respectively; obtaining a first threshold for the first function, and a second threshold for a combination of the first and second functions; applying the first function and the second function to each potential defect to obtain first and second scores, respectively; and determining a combined score of the first and second scores; and indicating as a defect of a potentially new type a potential defect when the first score is lower than the first threshold or the combined score exceeds the second threshold.
    Type: Application
    Filed: January 16, 2019
    Publication date: July 16, 2020
    Inventors: Kirill SAVCHENKO, Assaf ASBAG, Boaz COHEN
  • Publication number: 20190293669
    Abstract: An examination system, a method of obtaining a training set for a classifier, and a non-transitory computer readable medium, the method comprising: upon receiving in a memory device object inspection results comprising data indicative of potential defects, each potential defect of the potential defects associated with a multiplicity of attribute values defining a location of the potential defect in an attribute space: sampling by the processor a first set of defects from the potential defects, wherein the defects within the first set are dispersed independently of a density of the potential defects in the attribute space; and obtaining by the processor a training defect sample set comprising the first set of defects and data or parameters representative of the density of the potential defects in the attribute space.
    Type: Application
    Filed: March 22, 2018
    Publication date: September 26, 2019
    Inventors: Kirill SAVCHENKO, Assaf ASBAG, Boaz COHEN
  • Publication number: 20190096053
    Abstract: There are provided a classifier and a method of classifying defects in a semiconductor specimen. The classifier enables assigning each class to a classification group among three or more classification groups with different priorities. Classifier further enables setting purity, accuracy and/or extraction requirements separately for each class, and optimizing the classification results in accordance with per-class requirements. During training, the classifier is configured to generate a classification rule enabling the highest possible contribution of automated classification while meeting per-class quality requirements defined for each class.
    Type: Application
    Filed: September 28, 2017
    Publication date: March 28, 2019
    Inventors: Assaf ASBAG, Ohad SHAUBI, Kirill SAVCHENKO, Shiran GAN-OR, Boaz COHEN, Zeev ZOHAR