CAUSAL CONVOLUTION NETWORK FOR PROCESS CONTROL

Application No. 18/287,613

• Filed: Oct 19, 2023
• Priority: May 06, 2021 — EU 21172606.2
  • Jun 15, 2021 — EU 21179415.1
  • Apr 19, 2022 — nonprovisional of PCTEP2022060209
• Examiner: ALAM, MOHAMMED
• Assignee: ASML Holding N.V.
• OA Round: 1 (Non-Final)

Office Action

Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .

Non-Final Office Action

DETAILED ACTION

Examiner’s Notes
(a) Claim date: 10/19/2023.
(b) Priority date: 05/06/2021.

Claim Rejections - 35 USC 102

The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:A person shall be entitled to a patent unless:(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention.Claims 1–3, 10–11, 15–16, are rejected under 35 U.S.C. § 103 as being unpatentable over Roy (US 2021/0165399) in view of van-den (US 2018/0322891 A1).
(As to claim 1, 15, Roy in view of van-den discloses)
1. (Original) A method for configuring a semiconductor manufacturing process, the method comprising:

obtaining an input vector composed of a plurality of values of a first parameter associated with a semiconductor manufacturing process [Roy ¶0016, ¶0186–¶0188, Fig. 13 (1306, 1308): discloses obtaining a value of a high-order overlay parameter (D3y) based on measurements associated with a first operation of a process step and a first sampling scheme];

the plurality of values of the first parameter being based on respective measurements performed at a plurality of respective first times of operation of the semiconductor manufacturing process [ Roy further discloses that the recurrent neural network 1304 is trained and updated using a sequence of such measured parameter values obtained at successive lots/wafers L1, L3–L5, L6 (¶0187–¶0188, elements 1306, 1324, 626)];

using a causal convolution neural network to determine a predicted value of a second parameter at a second time of operation, no earlier than the latest of the first times of operation, based on the input vector; [Roy ¶0186–¶0188: discloses using a recurrent neural network (NN) 1304 to determine a predicted value 1312 of the high-order parameter at a subsequent operation L2 based on the measured value 1308 obtained at operation L1; van den ¶0030: discloses that a causal convolutional layer is a convolutional layer that, for each time step, generates an output that depends only on the inputs at the time step and at the time steps before that time step in the input sequence, i.e., and not on any inputs at any time steps after that time step, it would have been obvious to substitute van den Oord's causal convolutional neural network for Roy's recurrent neural network to perform the same predicted-value determination based on the input vector of first-parameter values];

PNG
media_image1.png
266
620
media_image1.png
Greyscale

configuring the semiconductor manufacturing process using an output of the causal convolution neural network [Roy Fig. 13, ¶0187–¶0188: discloses using the predicted value 1312 of the high-order parameter, determined by the neural network, to determine a control recipe 1314 of a subsequent operation L2 of the process step in the semiconductor manufacturing process, disclosing configuring the semiconductor manufacturing process using an output of the neural network, which as discussed above is rendered obvious to be a causal convolution neural network in view of van den Oord].

(As to claim 2, Roy in view of van den discloses)
2. (Original) The method of claim 1, in which the second time of operation is later than the first times of operation [Roy ¶0187–¶0188, Fig. 13: discloses that the predicted value of the high-order parameter is determined for operation L2, which is later than operation L1 at which the input measurement 1308 was taken, disclosing that the second time of operation is later than the first times of operation]].

(As to claim 3, 16, Roy in view of van den discloses)
3. (Original) The method of claim 1,

PNG
media_image2.png
506
380
media_image2.png
Greyscale

in which the causal convolution neural network comprises, in order, an input layer configured to receive the input vector, one or more convolutional layers, and an output layer configured to output the predicted value of the second parameter [van den ¶0024–¶0026, Fig. 1: discloses a neural network system 100 comprising a convolutional subnetwork 110 configured to receive a neural network input 102 (input vector) and, through one or more convolutional layers, generate an alternative representation 144, and an output layer 120 configured to process the alternative representation to generate an output (score distribution 146) that defines the predicted value for the time step, disclosing a causal convolution neural network comprising, in order, an input layer, one or more convolutional layers, and an output layer configured to output the predicted value]].

(As to claim 10, Roy in view of van den discloses)
10. (Original) The method of claim 1, in which the second parameter is the same as the first parameter [Roy ¶0186–¶0188, Fig. 13: discloses that the recurrent neural network predicts a subsequent value 1312/626 of the same high-order overlay parameter (the third-order scanner exposure magnification parameter D3y) whose earlier values 1306/1308/1328 were used as the measured input, disclosing that the second (predicted) parameter is the same as the first (measured) parameter]]

(As to claim 11, Roy in view of van den discloses)
11. (Original) The method of claim 10, in which the first values of the first parameter include first values obtained using a first sampling scheme, and the method further comprises using the predicted value of the first parameter to determine a control recipe of a subsequent operation of the process step in the semiconductor manufacturing process [Roy ¶0185–¶0188, Fig. 13: discloses that the value 1308 of the high-order parameter is obtained based on measurements 1306 using a spatially dense sampling scheme (a first sampling scheme), and that the predicted value 1312 of the parameter is used to determine a control recipe 1314 of a subsequent operation L2 of the process step in the semiconductor manufacturing process, disclosing using the predicted value of the first parameter to determine a control recipe of a subsequent operation of the process step]].

Allowable Subject Matter
The following claims would be allowable if all rejections/objections cited in this office action (if any) are overcome and rewritten to include all of the limitations of the base claim and any intervening claims.The reason for this allowance is: the claimed subject matter could not have been anticipated or obviated using any prior arts.Allowable claims are: 4-9, 12-14, 17-20.
Conclusion

The prior art made of record in the form PTO-892 are not relied upon is considered pertinent to applicant's disclosure.Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.Contact information:Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOHAMMED ALAM whose telephone number is [phone], email address: [[e-mail]] and fax number [phone]. The examiner can normally be reached on 10AM to 4PM (EST), Monday to Friday. If attempts to reach the examiner by telephone are unsuccessful, the Examiner's Supervisor, JACK CHIANG can be reached on [phone]. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300./Mohammed Alam/Primary Examiner, Art Unit 2851

Prosecution Timeline

• Oct 19, 2023: Application Filed
• Jul 21, 2026: Non-Final Rejection mailed — §102, §103 (current)