公开(公告)号：US10360477B2

公开(公告)日：2019-07-23

申请号：US15402169

申请日：2017-01-09

Applicant: KLA-Tencor Corporation

Inventor： Kris Bhaskar ,  Scott Young ,  Mark Roulo ,  Jing Zhang ,  Laurent Karsenti ,  Mohan Mahadevan ,  Bjorn Brauer

IPC: G06K9/00 ,  G06K9/62 ,  G06K9/66 ,  G06T7/00

Abstract: Methods and systems for performing one or more functions for a specimen using output simulated for the specimen are provided. One system includes one or more computer subsystems configured for acquiring output generated for a specimen by one or more detectors included in a tool configured to perform a process on the specimen. The system also includes one or more components executed by the one or more computer subsystems. The one or more components include a learning based model configured for performing one or more first functions using the acquired output as input to thereby generate simulated output for the specimen. The one or more computer subsystems are also configured for performing one or more second functions for the specimen using the simulated output.

公开(公告)号：US10346740B2

公开(公告)日：2019-07-09

申请号：US15609009

申请日：2017-05-31

Applicant: KLA-Tencor Corporation

Inventor： Jing Zhang ,  Kris Bhaskar

IPC: G06N7/02 ,  G06N7/04 ,  G06N7/06 ,  G06N7/08 ,  G06N3/08 ,  G06N3/04 ,  G06K9/62

Abstract: Methods and systems for training a neural network are provided. One system includes one or more components executed by one or more computer subsystems. The one or more components include a neural network configured for determining inverted features of input images in a training set for a specimen input to the neural network, a forward physical model configured for reconstructing the input images from the inverted features thereby generating a set of output images corresponding to the input images in the training set, and a residue layer configured for determining differences between the input images in the training set and their corresponding output images in the set. The one or more computer subsystems are configured for altering one or more parameters of the neural network based on the determined differences thereby training the neural network.

公开(公告)号：US10043261B2

公开(公告)日：2018-08-07

申请号：US15402094

申请日：2017-01-09

Applicant: KLA-Tencor Corporation

Inventor： Kris Bhaskar ,  Jing Zhang ,  Grace Hsiu-Ling Chen ,  Ashok Kulkarni ,  Laurent Karsenti

IPC: G06K9/00 ,  G06T7/00 ,  G03F1/84 ,  G03F1/86 ,  G03F7/20

Abstract: Methods and systems for generating simulated output for a specimen are provided. One method includes acquiring information for a specimen with one or more computer systems. The information includes at least one of an actual optical image of the specimen, an actual electron beam image of the specimen, and design data for the specimen. The method also includes inputting the information for the specimen into a learning based model. The learning based model is included in one or more components executed by the one or more computer systems. The learning based model is configured for mapping a triangular relationship between optical images, electron beam images, and design data, and the learning based model applies the triangular relationship to the input to thereby generate simulated images for the specimen.

公开(公告)号：US09916965B2

公开(公告)日：2018-03-13

申请号：US15394792

申请日：2016-12-29

Applicant: KLA-Tencor Corporation

Inventor： Kris Bhaskar ,  Grace Hsiu-Ling Chen ,  Keith Wells ,  Wayne McMillan ,  Jing Zhang ,  Scott Young ,  Brian Duffy

IPC: G01N21/00 ,  H01J37/22 ,  G01N21/95 ,  G01N23/225 ,  H01J37/06 ,  H01J37/28

CPC classification number: H01J37/222 ,  G01N21/9501 ,  G01N23/2251 ,  G01N2201/12 ,  G01N2223/304 ,  G01N2223/401 ,  G01N2223/418 ,  G01N2223/6116 ,  G01N2223/646 ,  G03F7/7065 ,  H01J37/06 ,  H01J37/226 ,  H01J37/28 ,  H01J2237/24475 ,  H01J2237/24495 ,  H01J2237/2817 ,  H01L22/20

Abstract: Hybrid inspectors are provided. One system includes computer subsystem(s) configured for receiving optical based output and electron beam based output generated for a specimen. The computer subsystem(s) include one or more virtual systems configured for performing one or more functions using at least some of the optical based output and the electron beam based output generated for the specimen. The system also includes one or more components executed by the computer subsystem(s), which include one or more models configured for performing one or more simulations for the specimen. The computer subsystem(s) are configured for detecting defects on the specimen based on at least two of the optical based output, the electron beam based output, results of the one or more functions, and results of the one or more simulations.

公开(公告)号：US20170351952A1

公开(公告)日：2017-12-07

申请号：US15609009

申请日：2017-05-31

Applicant: KLA-Tencor Corporation

Inventor： Jing Zhang ,  Kris Bhaskar

IPC: G06N3/08 ,  G06N3/04

CPC classification number: G06N3/08 ,  G06K9/6274 ,  G06N3/04 ,  G06N3/0454

Abstract: Methods and systems for training a neural network are provided. One system includes one or more components executed by one or more computer subsystems. The one or more components include a neural network configured for determining inverted features of input images in a training set for a specimen input to the neural network, a forward physical model configured for reconstructing the input images from the inverted features thereby generating a set of output images corresponding to the input images in the training set, and a residue layer configured for determining differences between the input images in the training set and their corresponding output images in the set. The one or more computer subsystems are configured for altering one or more parameters of the neural network based on the determined differences thereby training the neural network.

公开(公告)号：US20150012900A1

公开(公告)日：2015-01-08

申请号：US14321565

申请日：2014-07-01

Applicant: KLA-Tencor Corporation

Inventor： Eugene Shifrin ,  Ashok Kulkarni ,  Kris Bhaskar ,  Graham Michael Lynch ,  John Raymond Jordan, III ,  Chwen-Jiann Fang

IPC: H01L27/02 ,  G06F17/50 ,  G01N21/95

CPC classification number: G06F17/5081 ,  G01N21/9501 ,  G06F17/5045 ,  H01L22/12

Abstract: Systems and methods for detecting defects on a wafer are provided. One method includes determining locations of all instances of a weak geometry in a design for a wafer. The locations include random, aperiodic locations. The weak geometry includes one or more features that are more prone to defects than other features in the design. The method also includes scanning the wafer with a wafer inspection system to thereby generate output for the wafer with one or more detectors of the wafer inspection system. In addition, the method includes detecting detects in at least one instance of the weak geometry based on the output generated at two or more instances of the weak geometry in a single die on the wafer.

Abstract translation: 提供了用于检测晶片上的缺陷的系统和方法。 一种方法包括在晶片的设计中确定弱几何的所有实例的位置。 这些位置包括随机，不定期的位置。 弱几何包括与设计中的其他特征相比更容易出现缺陷的一个或多个特征。 该方法还包括用晶片检查系统扫描晶片，从而通过晶片检查系统的一个或多个检测器产生用于晶片的输出。 此外，该方法包括基于在晶片上的单个管芯中的弱几何形状的两个或多个实例处产生的输出来检测至少一个弱几何形状的检测。

公开(公告)号：US20140241610A1

公开(公告)日：2014-08-28

申请号：US14184417

申请日：2014-02-19

Applicant: KLA-Tencor Corporation

Inventor： Brian Duffy ,  Kris Bhaskar

IPC: G06T7/00

CPC classification number: G01N21/9501 ,  G06T7/50 ,  G06T7/62 ,  G06T2207/10016 ,  G06T2207/30252

Abstract: Generalized virtual inspectors are provided. One system includes two or more actual systems configured to perform one or more processes on specimen(s) while the specimen(s) are disposed within the actual systems. The system also includes one or more virtual systems coupled to the actual systems to thereby receive output generated by the actual systems and to send information to the actual systems. The virtual system(s) are configured to perform one or more functions using at least some of the output received from the actual systems. The virtual system(s) are not capable of having the specimen(s) disposed therein.

Abstract translation: 提供广泛的虚拟检查员。 一个系统包括两个或更多个实际系统，其被配置为在样本被布置在实际系统内时对样本执行一个或多个过程。 该系统还包括耦合到实际系统的一个或多个虚拟系统，从而接收由实际系统产生的输出并向实际系统发送信息。 虚拟系统被配置为使用从实际系统接收的至少一些输出来执行一个或多个功能。 虚拟系统不能将样本置于其中。

公开(公告)号：US20140199791A1

公开(公告)日：2014-07-17

申请号：US14083126

申请日：2013-11-18

Applicant: KLA-Tencor Corporation

Inventor： Allen Park ,  Ellis Chang ,  Michael Adel ,  Kris Bhaskar ,  Ady Levy ,  Amir Widmann ,  Mark Wagner ,  Songnian Rong

IPC: H01L21/66 ,  G06F17/50

CPC classification number: H01L22/12 ,  G06F17/5081

Abstract: Universal target based inspection drive metrology includes designing a plurality of universal metrology targets measurable with an inspection tool and measurable with a metrology tool, identifying a plurality of inspectable features within at least one die of a wafer using design data, disposing the plurality of universal targets within the at least one die of the wafer, each universal target being disposed at least proximate to one of the identified inspectable features, inspecting a region containing one or more of the universal targets with an inspection tool, identifying one or more anomalistic universal targets in the inspected region with an inspection tool and, responsive to the identification of one or more anomalistic universal targets in the inspected region, performing one or more metrology processes on the one or more anomalistic universal metrology targets with the metrology tool.

Abstract translation: 通用的基于目标的检测驱动度量包括设计多个通过检测工具测量的通用度量目标并且可以用计量工具测量，使用设计数据识别晶片的至少一个管芯内的多个检查特征，将多个通用目标 在晶片的至少一个模具内，每个通用目标被设置为至少接近所识别的可检查特征之一，用检查工具检查包含一个或多个通用目标的区域，以识别一个或多个异常通用目标 检查区域具有检查工具，并且响应于在被检查区域中识别一个或多个异常通用目标，对所述一个或多个异常通用度量目标与计量工具执行一个或多个计量过程。

公开(公告)号：US11580375B2

公开(公告)日：2023-02-14

申请号：US15394790

申请日：2016-12-29

Applicant: KLA-Tencor Corporation

Inventor： Kris Bhaskar ,  Laurent Karsenti ,  Scott Young ,  Mohan Mahadevan ,  Jing Zhang ,  Brian Duffy ,  Li He ,  Huajun Ying ,  Hung Nien ,  Sankar Venkataraman

IPC: G06N3/08 ,  G01Q30/02 ,  G06N3/067 ,  G06N20/00 ,  G06N3/04 ,  G06F30/20

Abstract: Methods and systems for accelerated training of a machine learning based model for semiconductor applications are provided. One method for training a machine learning based model includes acquiring information for non-nominal instances of specimen(s) on which a process is performed. The machine learning based model is configured for performing simulation(s) for the specimens. The machine learning based model is trained with only information for nominal instances of additional specimen(s). The method also includes re-training the machine learning based model with the information for the non-nominal instances of the specimen(s) thereby performing transfer learning of the information for the non-nominal instances of the specimen(s) to the machine learning based model.

公开(公告)号：US20190370955A1

公开(公告)日：2019-12-05

申请号：US16424431

申请日：2019-05-28

Applicant: KLA-Tencor Corporation

Inventor： Jing Zhang ,  Yujie Dong ,  Brian Duffy ,  Richard Wallingford ,  Michael Daino ,  Kris Bhaskar

IPC: G06T7/00 ,  G06N3/08 ,  G06N3/04 ,  G06K9/62 ,  G06N20/00 ,  G06F17/18 ,  G06N7/00

Abstract: Methods and systems for performing active learning for defect classifiers are provided. One system includes one or more computer subsystems configured for performing active learning for training a defect classifier. The active learning includes applying an acquisition function to data points for the specimen. The acquisition function selects one or more of the data points based on uncertainty estimations associated with the data points. The active learning also includes acquiring labels for the selected one or more data points and generating a set of labeled data that includes the selected one or more data points and the acquired labels. The computer subsystem(s) are also configured for training the defect classifier using the set of labeled data. The defect classifier is configured for classifying defects detected on the specimen using the images generated by the imaging subsystem.