DETAILED 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 .
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-2 and 10-13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cao (US 2020/0380362; IDS, 09/07/2023) in view of Guo (WO2022008174)
Cao discloses methods for training machine learning model for computation lithography. Cao explains that a neural network may be trained (i.e., whose parameters are determined) using a set of training data and the training data may comprise or consist of a set of training samples. (Para, 0075). Cao discloses each sample may be a pair comprising or consisting of an input object (typically a vector, which may be called a feature vector) and a desired output value (also called the supervisory signal). (Para, 0075). Cao discloses a training algorithm analyzes the training data and adjusts the behavior of the neural network by adjusting the parameters (e.g., weights of one or more layers) of the neural network based on the training data. (Para, 0075). Cao discloses the neural network after training can be used for mapping new samples. (Para, 0075).
Cao illustrates in Figure 8 a block diagram of a machine learning based architecture of a patterning process. (Para, 0097). Cao explains this block diagram illustrates different elements of the machine learning based architecture including (i) a set of trained machine learning models (e.g., 8004, 8006, 8008) representing, for example, a lithographic process, (ii) a machine learning model (e.g., 8002) representing or configured to predict mask patterns (e.g., a CTM image or OPC), and (iii) a cost function 8010 (e.g., a first cost function and a second cost function) used to trained different machine learning models according to the present disclosure. (Para, 0097). Cao discloses the machine learning architecture may be divided into several parts: (i) training of individual process model (e.g., 8004, 8006, and 8008), further discussed later in the disclosure, (ii) coupling the individual process models and further training and/or fine-tuning the trained process models based on a first training data set (e.g., printed patterns) and a first cost function (e.g., difference between printed patterns and predicted patterns), further discussed in Figure 9, and (iii) using the trained process models to train another machine learning model (e.g., 8002) configured to predict mask pattern (e.g., including OPC) based on a second training data set (e.g., a target pattern) and a second cost function (e.g., EPE between the target pattern and the predicted pattern), further discussed in Fig.10A. (Para, 0098). These disclosures teach and/or suggest the limitation of claims 2 and 13.
Cao discloses the patterning process may include the lithographic process which may be represented by one or more machine learning models such as convolutional neural networks (CNNs) or deep CNN. (Para, 0099). Cao explains each machine learning model (e.g., a deep CNN) may be individually pre-trained to predict an outcome of an aspect or process (e.g., mask diffraction, optics, resist, etching, etc.) of the patterning process. (Para, 0099). Cao discloses each such pre-trained machine learning model of the patterning process may be coupled together to represent the entire patterning process. (Para, 0099). Cao explains, as illustrated in Figure 8, a first trained machine learning model 8004 may be coupled to a second trained machine learning model 8006 and the second trained machine learning model 8006 may be further coupled to a third trained machine learning model 8008 such that the coupled models represent a lithographic process model. (Para, 0099). Cao discloses a fourth trained model (not illustrated) configured to predict an etching process may be coupled to the third trained model 8008, thus further extending the lithographic process model. (Para, 0099).
Cao explains cost function (e.g., the first cost function) may be defined based on a difference between the experimental data (i.e., printed patterns on a substrate) and the output of the third model 8008. (Para 0101). Cao explains the cost function may be a metric (e.g., RMS, MSE, MXE etc.) based on a parameter (e.g., CD, overlay) of the patterning process determined based on the output of the third trained model, for example, a trained resist CNN model that predicts an outcome of the resist process. (Para, 0101). Cao explains that during, the fine-tuning process, the training may involve modifying the parameters (e.g., weights, bias, etc.) of the process models so that the first cost function (e.g., the RMS) is reduced, in an embodiment, minimized. (Para, 0101). Cao discloses that consequently, the training and/or fine-tuning of the coupled models may generate a relatively more accurate model of the lithographic process compared to a non-fine-tuned model that is obtained by simply coupling individual trained models of different processes/aspects of the pattering process. (Para, 0101). These disclosures teach and/or suggest the limitation of claim 1, ‘A training method for a lithographic mask generation model, the method being performed by a computer device and the method comprising: …determining a model precision evaluation index according to the predictive mask map, and, the model precision evaluation index representing a mask prediction precision of the lithographic mask generation model…and adjusting at least one parameter of the lithographic mask generation model according to the training loss.’
Cao discloses the method for training a machine learning model configured to predict a mask pattern includes obtaining (i) a set of benchmark images, and (ii) a mask image corresponding to a target pattern, and training, by a hardware computer system, the machine learning model configured to predict the mask pattern based on the benchmark images and a cost function that determines a difference between the predicted mask pattern and the benchmark images. (Para, 0015). These disclosures teach and/or suggest the limitation of claim 10. Cao discloses in an embodiment, a first trained model 8004 may be a trained mask 3D CNN and/or a trained thin mask CNN model configured to predict a diffraction effect/behavior of a mask during the patterning process. (Para, 0102). Cao explains the mask may include a target pattern corrected for optical proximity corrections (e.g., SRAFs, Serifs, etc.) to enable printing of the target pattern on a substrate via the patterning process. (Para, 0102). Cao explains the first trained model 8004 may receive, for example, a continuous transmission mask (CTM) in the form of a pixelated image and based on the CTM image, the first trained model 8004 may predict a mask image (e.g., 640 in FIG. 6). (Para, 0102). Cao discloses the mask image may also be a pixelated image which may be further represented in a vector form, matrix form, tensor form, etc. for further processing by other trained models. (Para, 0102). Cao discloses, in an embodiment a deep convolutional neural network may be generated or a pre-trained model may be obtained. (Para, 0102). Cao explains that a first trained model 8004 to predict 3D mask diffraction may be trained as discussed earlier with respect to FIGS. 2-6 and this trained 3D CNN may then generate a mask image which can be sent to the second trained model 8006. (Para, 0102). These disclosures and the illustrations of Figures 6 and 8 teach and/or suggest the limitation of claim 1, ‘A training method for a lithographic mask generation model, the method being performed by a computer device and the method comprising: generating a predictive mask map corresponding to a chip layout through a lithographic mask generation model, the lithographic mask generation model being configured to generate a neural network model of the predictive mask map…’ and the limitation of claim 11.
Cao discloses the second trained model 8006 may be a trained CNN model configured to predict a behavior of projection optics (e.g., including an optical system) of a lithographic apparatus (also commonly referred as a scanner or a patterning apparatus). (Para, 0102). Cao explains, the second trained model may receive the mask image predicted by the first trained model 8004 and may predict an optical image or an aerial image. (Para, 0102). Cao discloses a second CNN model may be trained based on training data including a plurality of aerial images corresponding to a plurality of mask images, where each mask image may correspond to a selected pattern printed on the substrate. (Para, 0102). Cao discloses, the aerial images of the training data may be obtained from simulation of optical model. (Para, 0102) Cao explains that based on the training data, the weights of the second CNN model may be iteratively adjusted such that a cost function is reduced, in an embodiment, minimized. (Para, 0102). Cao discloses after several iterations, the cost function may converge (i.e., no further improvement in predicted aerial image is observed) at which point the second CNN model may be considered as the second trained model 8006. (Para, 0102). These disclosures teach and/or suggest the limitation of claim 1, ‘A training method for a lithographic mask generation model, the method being performed by a computer device and the method comprising: …generating a predictive wafer pattern corresponding to the predictive mask map through a pre-trained wafer pattern generation model, the wafer pattern generation model being a machine learning model constructed based on a neural network…’
Cao discloses the third trained model 8008 may be a CNN model configured to predict a behavior of a resist process, as discussed earlier. (Para, 0105). Cao discloses the training of a machine learning model (e.g., a ML-resist model) is based on (i) an aerial image(s), for example, predicted by an aerial image model (e.g., a machine learning based model or physics based model), and/or (ii) a target pattern (e.g., a mask image rendered from target layout). Further, the training process may involve reducing (in an embodiment, minimize), a cost function that describes the difference between a predicted resist image and an experimentally measured resist image (SEM image). (Para, 0105). Cao explains the cost function can be based on image pixel intensity difference, contour to contour difference, or CD difference, etc. (Para, 0105). Cao discloses, after the training, the ML-resist model can predict a resist image from an input image, for example, an aerial image. (Para, 0106). Cao also discloses as part of this machine learning model, there is provided a method for training a machine learning model configured to predict mask rule check violations of a mask pattern. (Para, 0017). Cao explains the method includes obtaining (i) a set of mask rule check, (ii) a set of mask patterns, and training, by a hardware computer system, the machine learning model configured to predict mask rule check violations based on the set of mask rule check, the set of mask patterns, and a cost function based on a mask rule check metric, wherein the cost function is a difference between the predicted mask rule check metric and a truth mask rule check metric. (Para, 0017). These disclosures teach and/or suggest the limitation of claim 1, ‘A training method for a lithographic mask generation model, the method being performed by a computer device and the method comprising: … determining a mask quality evaluation index according to the predictive wafer pattern, the mask quality evaluation index representing a quality of the predictive mask map…’ and the limitation of claim 2.
Cao also discloses for the methods disclosed, there is provided computer program product comprising a non-transitory computer readable medium having instructions recorded thereon, the instructions when executed by a computer implementing any of the methods above. (Para, 0021). This disclosure teaches and/or suggests the limitations of claim 12, ‘ A non-transitory computer readable medium, storing one or more programs, the one or more programs being configured to be executed by at least one processor to cause a computer to perform steps comprising: generating a predictive mask map corresponding to a chip layout through a lithographic mask generation model, the lithographic mask generation model being configured to generate a neural network model of the predictive mask map; generating a predictive wafer pattern corresponding to the predictive mask map through a pre-trained wafer pattern generation model, the wafer pattern generation model being a machine learning model constructed based on a neural network; determining a model precision evaluation index according to the predictive mask map, and, the model precision evaluation index representing a mask prediction precision of the lithographic mask generation model; determining a mask quality evaluation index according to the predictive wafer pattern, the mask quality evaluation index representing a quality of the predictive mask map… and adjusting at least one parameter of the lithographic mask generation model according to the training loss.’
Still, the disclosures of Cao as discussed above fail to teach and/or suggest the limitation of claim 1, ‘ A training method for a lithographic mask generation model, the method being performed by a computer device and the method comprising: … determining a training loss according to the model precision evaluation index and the mask quality evaluation index…’ as well as the limitation of claim 12, ‘ A non-transitory computer readable medium, storing one or more programs, the one or more programs being configured to be executed by at least one processor to cause a computer to perform steps comprising: …determining a training loss according to the model precision evaluation index and the mask quality evaluation index…’ However, the disclosures of Cao in view of the disclosures of Guo provide such teachings.
Guo is also directed to controlling a patterning process using machine learning. Guo discloses a hybrid cost function that reduces model error by comparing the machine learning model of various elements of the patterning process. (Para, 0160-0164). Guo explains this allows for more precise fine tuning of the machine learning model of the patterning process. (Para, 0160). Guo discloses for each of machine learning model which comprises the patterning process, the training date generated via simulation is compared to the actual date of the patterning process. (Para, 0160), which helps to fine tune each individual model. (Para, 0160). Guo discloses cost function can be further fine tuned for the entire process by reducing error for using the hybrid cost function disclosed. (Para, 0163). Guo discloses an exemplary equation comparing two models
PNG
media_image1.png
56
565
media_image1.png
Greyscale
where Cost mu_tau (Zn_ m) is a cost associated with the predicted mu tau values, w is a weighting term, and Cost LHFF (Zn_ m) is a cost associated with the lens heating feedforward control signal. (Para, 0163).
It would have been obvious to one of ordinary skill in the art at the time of filing of the present application by Applicant to modify the disclosures of Cao in view of the disclosures of Guo because both are directed to analogous methods of machine learning for training models of lithographic processes and Guo provides a more comprehensive calculation for training loss that can be applied as the cost function for determining training loss and also minimizing training loss in the machine learning model disclosed in Cao.
Allowable Subject Matter
Claims 3-9 and 14-20 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter: The disclosures of Cao in view of Guo as discussed above fail to teach and/or suggest the limitation of claim 3, ‘ The method according to claim 1, further comprising: acquiring a complexity evaluation index corresponding to the predictive mask map, the complexity evaluation index being representing a complexity of the predictive mask map, wherein determining the training loss according to the model precision evaluation index and the mask quality evaluation index comprises determining the training loss according to the model precision evaluation index, the mask quality evaluation index, and the complexity evaluation index.’ Moreover, the disclosures of Cao in view of Guo as discussed above fail to teach and/or suggest the limitation of claim 14, ‘ The non-transitory computer readable medium according to claim 12, wherein the one or more programs are configured to be executed by the at least one processor to cause the computer to perform a step comprising: acquiring a complexity evaluation index corresponding to the predictive mask map, the complexity evaluation index being representing a complexity of the predictive mask map, wherein determining the training loss according to the model precision evaluation index and the mask quality evaluation index comprises determining the training loss according to the model precision evaluation index, the mask quality evaluation index, and the complexity evaluation index.’ The prior art fails to provide other relevant disclosures which are properly combinable with Cao and/or Guo to teach and/or suggest the limitations of claims 3 and/or 14. Claims 4-9 and 15-20 respectively depend directly from claims 3 and 14 respectively. Therefore, claims 3-9 and 14-20 include allowable subject matter.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CALEEN O SULLIVAN whose telephone number is (571)272-6569. The examiner can normally be reached Mon-Fri: 7:30 am-4:00 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Dale Page can be reached at 571-270-7877. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/CALEEN O SULLIVAN/Primary Examiner, Art Unit 2899
Prosecution Insights