Optimization based on machine learning

• US 10,409,165 B2
• Filed: 11/18/2015
• Issued: 09/10/2019
• Est. Priority Date: 12/15/2014
• Status: Active Grant

• Current Assignee: ASML Netherlands BV (ASML Holding NV)
• Original Assignee: ASML Netherlands BV (ASML Holding NV)
• Inventors: Liu, Xiaofeng
• Primary Examiner(s): Doan, Nghia M
• Application Number:
  US15/528,034
• Publication Number:
  US 20180314163A1
• Time in Patent Office: 1,392 Days
• Field of Search: 716 50- 56, 430 5, 430 30
• CPC Class Codes: G03F 7/70125 Use of illumination settings tailored to particular mask patterns details of setting means G03F7/70091
• CPC Class Codes: G03F 7/70441 Optical proximity correction [OPC]
• CPC Class Codes: G03F 7/70491 Information management, e.g. software; Active and passive control, e.g. details of controlling exposure processes or exposure tool monitoring processes
• CPC Class Codes: G06F 30/00 Computer-aided design [CAD]
• CPC Class Codes: G06F 30/20 Design optimisation, verification or simulation optimisation, verification or simulation of circuit designs G06F30/30
• CPC Class Codes: G06F 30/27 using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]

Abstract

A method for improving a lithographic process for imaging a portion of a design layout onto a substrate using a lithographic apparatus, the method including: obtaining a first source of the lithographic apparatus; classifying the first source into a class among a plurality of possible classes, based on one or more numerical characteristics of the first source, using a machine learning model, by a computer; determining whether the class is among one or more predetermined classes; only when the class is among the one or more predetermined classes, adjusting one or more source design variables to obtain a second source.

Claims

1. A method for improving imaging a portion of a design layout onto a substrate using a lithographic apparatus, the method comprising:

classifying, by a hardware computer system and a machine learning model, a first illumination source of the lithographic apparatus into a class among a plurality of possible classes, based on one or more numerical characteristics of the first illumination source, the machine learning model developed on illumination source numerical characteristics in order to classify illumination sources;

determining that the class of the first illumination source is among one or more predetermined classes of the plurality of possible classes; and

only when the class of the first illumination source is among the one or more predetermined classes, adjusting one or more illumination source design variables to obtain a second illumination source.

2. The method of claim 1, further comprising, only when the class of the first illumination source is among the one or more predetermined classes, adjusting one or more non-illumination source design variables.

3. The method of claim 1, wherein the machine learning model is supervised.

4. The method of claim 1, wherein the machine learning model is selected from:

Decision trees, Ensembles (Bagging, Boosting, Random forest), k-NN, Linear regression, Naive Bayes, Neural networks, Logistic regression, Perceptron, Support vector machine (SVM), Relevance vector machine (RVM), and/or deep learning.

5. The method of claim 1, wherein the machine learning model is non-probabilistic, or wherein the machine learning model is a binary non-linear classifier.

6. The method of claim 1, wherein the machine learning model uses a kernel method.

7. The method of claim 1, wherein the machine learning model is trained with a training set comprising elements of feature vectors and classes of the feature vectors, wherein the feature vectors comprise one or more numerical characteristics of an illumination source of the lithographic apparatus.

8. The method of claim 1, wherein the plurality of possible classes consists of two classes and a class is determined from either of the two classes of the plurality of possible classes.

9. The method of claim 1, wherein the first illumination source is obtained by randomly selecting a point in a space of the one or more illumination source design variables, or wherein the first illumination source is obtained by a Monte-Carlo method, or wherein in the first illumination source is obtained from optimizing an illumination source of the lithographic apparatus by adjusting one or more illumination source design variables.

10. The method of claim 1, further comprising obtaining the first illumination source by optimizing an illumination source of the lithographic apparatus.

11. The method of claim 10, wherein the illumination source is optimized simultaneously with the portion of the design layout.

12. The method of claim 1, further comprising generating an element with the second illumination source.

13. The method of claim 12, further comprising updating a training set with the element to obtain an updated training set.

14. The method of claim 13, further comprising training the machine learning model with the updated training set.

15. A computer program product comprising a non-transitory computer readable medium having instructions recorded thereon, the instructions, upon execution by a hardware computer system, configured to cause the hardware computer system to at least:

classify, by a machine learning model, a first illumination source of the lithographic apparatus into a class among a plurality of possible classes, based on one or more numerical characteristics of the first illumination source, the machine learning model developed on illumination source numerical characteristics in order to classify illumination sources;

determine whether the class of the first illumination source is among one or more predetermined classes of the plurality of possible classes; and

only when the class of the first illumination source is among the one or more predetermined classes, adjust one or more illumination source design variables to obtain a second illumination source.

16. The computer program product of claim 15, wherein instructions are further configured to, only when the class of the first illumination source is among the one or more predetermined classes, adjust one or more non-illumination source design variables.

17. The computer program product of claim 15, wherein the machine learning model is supervised.

18. The computer program product of claim 15, wherein the machine learning model is non-probabilistic, or wherein the machine learning model is a binary non-linear classifier, or wherein the machine learning model uses a kernel method.

19. The computer program product of claim 15, wherein the machine learning model is trained with a training set comprising elements of feature vectors and classes of the feature vectors, wherein the feature vectors comprise one or more numerical characteristics of an illumination source of the lithographic apparatus.

20. The computer program product of claim 15, wherein the plurality of possible classes consists of two classes and a class is determined from either of the two classes of the plurality of possible classes.