CTNF 18/471,661 CTNF 82888 DETAILED ACTION Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. DETAILED ACTION The instant application having Application No. 18471661 has a total of 20 claims pending in the application, all of which are ready for examination by the examiner. I. ACKNOWLEDGEMENT OF REFERENCES CITED BY APPLICANT Information Disclosure Statement As required by M.P.E.P 609(c), the applicant’s submissions of the Information Disclosure Statements dated 9/21/23, 9/27/23, 12/17/249/23/25 and 4/8/26 are acknowledged by the examiner and the cited references have been considered in the examination of the claims now pending. As required by M.P.E.P 609 C(2), a copy of the PTOL-1449 initialed and dated by the examiner is attached to the instant office action. II. REJECTIONS NOT BASED ON PRIOR ART Claim Rejections – 35 USC § 101 07-04-01 AIA 07-04 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Step 1: Claim 1 is a process type claim. Claim 12 is a machine type claim. Claim 18 is a manufacture type claim. Therefore, claims 1-X are directed to either a process, machine, manufacture or composition of matter. As per claim 1 , 2A Prong 1 : “ generating a first embedding for the first input” A user mentally or with pencil and paper reformats the data as needed to fit their model . “Generating a second embedding for the second input” The user mentally or with pencil and paper reforms the data s needed to fit their model . “aggregating the first and second embeddings to generate a combined embedding” The user merges the two sets of data . “applying a transformation function to the combined embedding to generate a transformed combined embedding” The user mentally or with pencil and paper manipulates the data mathematically to transform it as needed by their model . “processing the transformed combined embedding … to extract a first transformed embedding for the first input and a second transformed embedding for the second input” The user mentally or with pencil and paper separates out the data again . “applying an inference function to the first transformed embedding to generate a first output” The user mentally or with pencil and paper makes a prediction based upon the data. “applying the inference function to the second transformed embedding to generate a second output” The user mentally or with pencil and paper makes a prediction based upon the data . 2A Prong 2 : This judicial exception is not integrated into a practical application. Additional elements: “a binding network”, “the binding network”, “an unbinding network” (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: Claims contain nothing more than generic machine learning networks with no additional detail or limitations than make them anything more than generic, off the shelf machine learning networks. “accessing a first input and a second input” (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)). 2B : The claim does not include additional elements individually or in combination that are sufficient to amount to significantly more than the judicial exception. Additional elements: “a binding network”, “the binding network”, “an unbinding network” (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: Claims contain nothing more than generic machine learning networks with no additional detail or limitations than make them anything more than generic, off the shelf machine learning networks. “accessing a first input and a second input” (MPEP 2106.05(d)(II) indicate that merely “receiving and transmitting data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed accessing step is well-understood, routine, conventional activity is supported under Berkheimer). As per claims 2-9 and 11 , these claims contain additional mental steps and generic machine learning models, and are rejected for similar reasons to claim 1. As per claim 10 , this claim contains additional mental steps similar to claim 1 and is rejected for similar reasons. As per claim 12 , 2A Prong 1 : “ generate a first embedding for the first input” A user mentally or with pencil and paper reformats the data as needed to fit their model . “Generate a second embedding for the second input” The user mentally or with pencil and paper reforms the data as needed to fit their model . “aggregate the first and second embeddings to generate a combined embedding” The user merges the two sets of data . “apply a transformation function to the combined embedding to generate a transformed combined embedding” The user mentally or with pencil and paper manipulates the data mathematically to transform it as needed by their model . “process the transformed combined embedding … to extract a first transformed embedding for the first input and a second transformed embedding for the second input” The user mentally or with pencil and paper separates out the data again . “apply an inference function to the first transformed embedding to generate a first output” The user mentally or with pencil and paper makes a prediction based upon the data. “apply the inference function to the second transformed embedding to generate a second output” The user mentally or with pencil and paper makes a prediction based upon the data . 2A Prong 2 : This judicial exception is not integrated into a practical application. Additional elements: ”one or more memories”, “one or more processors” (mere instructions to apply the exception using a generic computer component); “a binding network”, “the binding network”, “an unbinding network” (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: Claims contain nothing more than generic machine learning networks with no additional detail or limitations than make them anything more than generic, off the shelf machine learning networks. “accessing a first input and a second input” (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)). 2B : The claim does not include additional elements individually or in combination that are sufficient to amount to significantly more than the judicial exception. Additional elements: ”one or more memories”, “one or more processors” (mere instructions to apply the exception using a generic computer component) “a binding network”, “the binding network”, “an unbinding network” (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: Claims contain nothing more than generic machine learning networks with no additional detail or limitations than make them anything more than generic, off the shelf machine learning networks. “accessing a first input and a second input” (MPEP 2106.05(d)(II) indicate that merely “receiving and transmitting data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed access step is well-understood, routine, conventional activity is supported under Berkheimer). As per claims 13-16 , these claims contain additional mental steps and generic machine learning models, and are rejected for similar reasons to claim 1. As per claim 17 , this claim contains additional mental steps similar to claim 1 and is rejected for similar reasons. As per claim 18 , 2A Prong 1 : “ generate a first embedding for the first input” A user mentally or with pencil and paper reformats the data as needed to fit their model . “Generate a second embedding for the second input” The user mentally or with pencil and paper reforms the data as needed to fit their model . “aggregate the first and second embeddings to generate a combined embedding” The user merges the two sets of data . “apply a transformation function to the combined embedding to generate a transformed combined embedding” The user mentally or with pencil and paper manipulates the data mathematically to transform it as needed by their model . “process the transformed combined embedding … to extract a first transformed embedding for the first input and a second transformed embedding for the second input” The user mentally or with pencil and paper separates out the data again . “apply an inference function to the first transformed embedding to generate a first output” The user mentally or with pencil and paper makes a prediction based upon the data. “apply the inference function to the second transformed embedding to generate a second output” The user mentally or with pencil and paper makes a prediction based upon the data . 2A Prong 2 : This judicial exception is not integrated into a practical application. Additional elements: “Computer readable storage medium” “one or more processors” (mere instructions to apply the exception using a generic computer component); “a binding network”, “the binding network”, “an unbinding network” (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: Claims contain nothing more than generic machine learning networks with no additional detail or limitations than make them anything more than generic, off the shelf machine learning networks. “accessing a first input and a second input” (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)). 2B : The claim does not include additional elements individually or in combination that are sufficient to amount to significantly more than the judicial exception. Additional elements: “Computer readable storage medium” “one or more processors” (mere instructions to apply the exception using a generic computer component) “a binding network”, “the binding network”, “an unbinding network” (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: Claims contain nothing more than generic machine learning networks with no additional detail or limitations than make them anything more than generic, off the shelf machine learning networks. “accessing a first input and a second input” (MPEP 2106.05(d)(II) indicate that merely “receiving and transmitting data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed access step is well-understood, routine, conventional activity is supported under Berkheimer). As per claims 19-20 , these claims contain additional mental steps and generic machine learning models, and are rejected for similar reasons to claim 1. Claim Rejections - 35 USC § 112 07-30-02 AIA The following is a quotation of 35 U.S.C. 112(b): (b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention. The following is a quotation of 35 U.S.C. 112 (pre-AIA), second paragraph: The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention. 07-34-01 Claim 9 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. As per claim 9 , this claim calls for “weights of the more convolutional layers are trained using isometric regularization.” The use of the term “more” is a relative term, and the Examiner is unsure how a convolutional layer can be “more convolutional” than another convolutional layer. Since the applicant has not made a comparison or explanation on how one convolutional lay can be more convolutional than another, this causes the claim to be confusing and therefore rejected under U.S.C. 112(b) for failing to particularly point out and claim the intended invention. III. REJECTIONS BASED ON PRIOR ART Examiners Note : Some rejections will be followed by an ‘EN’ that will denote an examiners note. This will be placed to further explain a rejection. Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. 07-20-aia AIA 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. 07-21-aia AIA Claim s 1-3, 5-8, 12-14, 16, and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Alam et al (“Deploying Convolution Networks on Untrusted Platforms Using 2D Holographic Reduced Representations”) in view of Ganesan et al (“learning with Holographic Reduced Representations”) . As per claims 1, 12, and 18 , Alam discloses, “A method comprising” (Abstract; EN; this denotes the system running on hardware and other computer environments, which will include some form of processor and memory) . “Accessing a first input and a second input” (Pg.4, particularly Figure 1 and associated paragraphs; EN: this denotes multiple inputs) . “generating a first embedding for the first input using a binding network” (Pg.4, figure 1 and associated paragraphs; EN: The orange box on the left denotes the binding network which performs the binding) . “generating a second embedding fort the second input using the binding network” (Pg.4, figure 1 and associated paragraphs; EN: The orange box on the left denotes the binding network which performs the binding) . “applying a transformation function to the … embedding to generated a transformed … embedding” (Pg.3-4, particularly section 4.2; Figure 1 and associated paragraphs; EN: this denotes the UNet main network, which takes the bound information and “transforms” it into an output in the main neural network) . “processing the transformed … embedding, using an unbinding network…” (Pg.4, particularly C1, last paragraph; C2 first paragraph and Figure 2 and associated paragraphs; EN: this denotes the use of an unbinding network to recover the original data and use it for predictions) . “applying an inference function to the … transformed bedding to generate a first output” (Pg.4, particularly C1, last paragraph; C2 first paragraph and Figure 2 and associated paragraphs; EN: this denotes the use of an unbinding network to recover the original data and use it for predictions) . “applying the inference function to the … transformed embedding to generate a second output” (Pg.4, particularly C1, last paragraph; C2 first paragraph and Figure 2 and associated paragraphs; EN: this denotes the use of an unbinding network to recover the original data and use it for predictions) . However, Alam fails to explicitly disclose, “aggregating the first and second embeddings to generate a combined embedding”, “combined embedding”, “to extract a first transformed embedding for the first input and a second transformed embedding for the second input”, “First transformed embedding”, and “second transformed embedding.” Ganesan discloses, “aggregating the first and second embeddings to generate a combined embedding”, “combined embedding” (pg.3-4, particularly section 3; EN: this denotes adding together different bound together components) . “to extract a first transformed embedding for the first input and a second transformed embedding for the second input”, First transformed embedding”, and “second transformed embedding” (pg.3-4, particularly section 3; EN: this denotes unbinding the data from the bound together data as needed, allowing individual elements of the combined data to be pulled out for use by the network) . Alam and Ganesan are analogous art because both involve Holographic reduced representations (HRR) in neural networks. Before the effective filing date it would have been obvious to one skilled in the art of neural network HRR to combine the work of Alam and Ganesan in order to allow multiple inputs to be combined together and then separated out for use. The motivation for doing so would be because “the binding is also distributive (and associative) … This allows us to construct complex symbolic relationships …” Ganesan, Pg.4, first paragraph) and the reference goes on to show in that paragraph that individual elements can be pulled from the complex relationships as needed. When combined with the Alam reference, this denotes pulling out whatever data from the combined convolutions as needed by the system and using them. Therefore before the effective filing date it would have been obvious to one skilled in the art of neural network HRR to combine the work of Alam and Ganesan in order to allow multiple inputs to be combined together and then separated out for use. As per claims 2, 13, and 19 , Alam discloses, “wherein the first embedding is generated by combining a first binding key and the first input using the binding network” (Pg.4, particularly figure 4 and associated paragraphs; EN: this denotes the use of secrets to bind to the inputs, with the secrets being the binding key) . As per claims 3, 14, and 20 , Alam discloses, “wherein the binding network comprises: a convolutional layer to generate a convolution output based on the first input” (Pg.3, particularly section 3; EN: this denotes binding using circular convolution) . “a binding operator to generate the first embedding by performing circular convolution on the first input using the binding key” (Pg.4, particularly figure 4 and associated paragraphs; EN: this denotes the use of secrets to bind to the inputs, with the secrets being the binding key) . As per claims 5 and 16 , Alam discloses, “wherein extracting the … transformed embedding for the first input comprises processing a first unbinding key and the transformed combined embedding using the unbinding network” (Pg.4, particularly Figure 1 and associated paragraphs; C1, last paragraph; C2, first paragraph; EN: this denotes the prediction network which unbinds the data using the secret key) . Ganesan discloses, “extracting the first transformed embedding” (pg.3-4, particularly section 3; EN: this denotes unbinding the data from the bound together data as needed, allowing individual elements of the combined data to be pulled out for use by the network) . As per claims 6 , Alam discloses, “wherein the first unbinding key comprises a trained value learned while the transformation function was trained” (Pg.3, particularly section 4; EN: This denotes the secret being part of the training process) . As per claim 7 , Alam discloses, “Wherein the first binding key comprises a random value assigned while the transformation function was trained” (Pg.3, particularly C2, section 4.1; EN: this denotes the secret being random) . As per claim 8 , Alam discloses, “wherein the transformation function comprises one or more convolutional layers” (Pg.3, particularly C2, section 4.2; EN: this denotes the main network, U-net, being a convolutional neural network) . Alam fails to explicitly disclose “and one or more non-linear activation functions.” While Alam fails to explicitly disclose, “and one or more non-linear activation functions” the Examiner takes Official Notice that the use of nonlinear activation functions, such as ReLU would be obvious to one of ordinary skill in the art of neural networks at the time of filing, as nonlinear activation functions allow neural networks to show improved performance over other activation functions including increased accuracy and removal of the vanishing gradient problem . Claim Rejections - 35 USC § 103 07-21-aia AIA Claim s 4 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Alam et al (“Deploying Convolution Networks on Untrusted Platforms Using 2D Holographic Reduced Representations”) and Ganesan et al (“learning with Holographic Reduced Representations”) further in view of Plate (“Holographic Reduced Representations”) . As per claims 4 and 15 , Alam discloses, “wherein the unbinding network comprises an unbinding operator to generate a convolution output based on the first input to generate an unbound … ” (Pg.4, particularly Figure 1 and associated paragraphs; C1, last paragraph; C2, first paragraph; EN: this denotes the prediction network which unbinds the data) . “based on performing circular…” on the transformed combined embedding using an unbinding key” (Pg.4, particularly Figure 1 and associated paragraphs; C1, last paragraph; C2, first paragraph; EN: this denotes the prediction network which unbinds the data using the secret key) . However, Alam fails to explicitly disclose “circular correlation” and “tensor.” Plate discloses, “circular correlation” (Pg.626, particularly C1, first paragraph; Figure 4, Figure 5; EN: this denotes that the inverse of circular convolution is the circular correlation, when combined with Alam, this denotes the inverse of the circular convolution to be circular correlation) . “tensor” (Pg.625, particularly C1, third paragraph; EN: this denotes the use of tensors with the HRR) . Alam and Plate are analogous art because both involve Holographic reduced representations (HRR) in neural networks. Before the effective filing date it would have been obvious to one skilled in the art of neural network HRR to combine the work of Alam and Plate in order to the use of circular correlation and the use of tensors with HRR networks. The motivation for using circular correlation would be because “While the circular convolution operation is straight forward, what is remarkable is the circular correlation … is an approximate inverse operation of it” (Pg.626, C1, first paragraph) or in the case of Alam, allow the use of convolutional correlation for the inverse operations described in Alam in order to unbind the information as needed. The motivation for tensor use would be to solve “the problem of expanding dimensionality [to be] tackled in a number of ways …” (Pg.625, C1, third paragraph). Therefore before the effective filing date it would have been obvious to one skilled in the art of neural network HRR to combine the work of Alam and Plate in order to the use of circular correlation and the use of tensors with HRR networks . Claim Rejections - 35 USC § 103 07-21-aia AIA Claim s 9 is rejected under 35 U.S.C. 103 as being unpatentable over Alam et al (“Deploying Convolution Networks on Untrusted Platforms Using 2D Holographic Reduced Representations”) and Ganesan et al (“learning with Holographic Reduced Representations”) further in view of Beshkov et al (“Isometric Representation in Neural Networks Improve Robustness”) . Alam fails to explicitly disclose, “wherein, during training, weights of the more convolutional layers are trained using isometric regularization.” Beshkov discloses, “wherein, during training, weights of the more convolutional layers are trained using isometric regularization” (abstract; EN: this denotes using isometric regularization terms during training in order to improve robustness against adversarial attacks) . Alam and Beshkov are analogous art because both involve neural networks. Before the effective filing date it would have been obvious to one skilled in the art of neural networks to combine the work of Alam and Beshkov in order to use isometric regularization. The motivation for doing so would be because “isometric regularization term improves the robustness to adversarial attacks MNIST” (Beshkov, Abstract) or in the case of Alam, allow the system to be more robust against outside attacks against the system. Therefore before the effective filing date it would have been obvious to one skilled in the art of neural networks to combine the work of Alam and Beshkov in order to use isometric regularization . Claim Rejections - 35 USC § 103 07-21-aia AIA Claim s 10 is rejected under 35 U.S.C. 103 as being unpatentable over Alam et al (“Deploying Convolution Networks on Untrusted Platforms Using 2D Holographic Reduced Representations”) and Ganesan et al (“learning with Holographic Reduced Representations”) further in view of Cohen et al (US 20210027083 A1) . Ganesan discloses, “wherein aggregating the first and second embeddings to form a combined embedding comprises” (pg.3-4, particularly section 3; EN: this denotes adding together different bound together components) . “Combining the … elements to generate the combined embedding” (pg.3-4, particularly section 3; EN: this denotes adding together different bound together components) . However, Alam and Ganesan fail to explicitly disclose, “applying one or more masks to the first and second embeddings to extract one or more elements from the first and second embeddings”, and “the extracted elements.” Cohen discloses, “applying one or more masks to the first and second embeddings to extract one or more elements from the first and second embeddings”, and “the extracted elements” (pg.5 particularly paragraph 0063; EN: this denotes the use of a selection mask to select aspects from images, such as the images of Alam, as needed) . Cohen and Alam modified by Ganesan are analogous art because both involve image processing. Before the effective filing date it would have been obvious to one skilled in the art of image processing to combine the work of Cohen and Alam modified by Ganesan in order to use masks to select data. The motivation for doing so would be to “generate an object mask for the detected query object. For example, the object selection system 106 can provide the detected object to an object mask neural network, which generates an object mask (e.g., selection mask) for the object” (Cohen, Pg.5, paragraph 0063) or in the case of Alam modified by Ganesan, allow the system to get the aspects of the image data they are interested in via masks as needed by the system. Therefore before the effective filing date it would have been obvious to one skilled in the art of image processing to combine the work of Cohen and Alam modified by Ganesan in order to use masks to select data. Allowable Subject Matter Claims 11 and 17 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. Claims 11 and 17 disclose a third input which has a third embedding for the input using the binding network, then creating a fourth embedding for the third input using the binding network by reprocessing the third input a second time by the binding network then combining the embeddings generated by these two processes into a combined embedding. While the prior art discloses the binding network as shown above and the combining of previously bound data into aggregated embeddings, then splitting them up and the prior art does not disclose processing the same input multiple times to create two embeddings and then combining those two embeddings together, unbinding them later, then averaging them in order to prepare an inference on the averaged embedding. Conclusion The examiner requests, in response to this Office action, support be shown for language added to any original claims on amendment and any new claims. That is, indicate support for newly added claim language by specifically pointing to page(s) and line no(s) in the specification and/or drawing figure(s). This will assist the examiner in prosecuting the application. When responding to this office action, Applicant is advised to clearly point out the patentable novelty which he or she thinks the claims present, in view of the state of the art disclosed by the references cited or the objections made. He or she must also show how the amendments avoid such references or objections See 37 CFR 1.111(c). Any inquiry concerning this communication or earlier communications from the examiner should be directed to BEN M RIFKIN whose telephone number is (571)272-9768. The examiner can normally be reached Monday-Friday 9 am - 5 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, Alexey Shmatov can be reached at (571) 270-3428. 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. /BEN M RIFKIN/Primary Examiner, Art Unit 2123 Application/Control Number: 18/471,661 Page 2 Art Unit: 2123 Application/Control Number: 18/471,661 Page 3 Art Unit: 2123 Application/Control Number: 18/471,661 Page 4 Art Unit: 2123 Application/Control Number: 18/471,661 Page 5 Art Unit: 2123 Application/Control Number: 18/471,661 Page 6 Art Unit: 2123 Application/Control Number: 18/471,661 Page 7 Art Unit: 2123 Application/Control Number: 18/471,661 Page 8 Art Unit: 2123 Application/Control Number: 18/471,661 Page 9 Art Unit: 2123 Application/Control Number: 18/471,661 Page 10 Art Unit: 2123 Application/Control Number: 18/471,661 Page 11 Art Unit: 2123 Application/Control Number: 18/471,661 Page 12 Art Unit: 2123 Application/Control Number: 18/471,661 Page 13 Art Unit: 2123 Application/Control Number: 18/471,661 Page 14 Art Unit: 2123 Application/Control Number: 18/471,661 Page 15 Art Unit: 2123 Application/Control Number: 18/471,661 Page 16 Art Unit: 2123 Application/Control Number: 18/471,661 Page 17 Art Unit: 2123 Application/Control Number: 18/471,661 Page 18 Art Unit: 2123 Application/Control Number: 18/471,661 Page 19 Art Unit: 2123 Application/Control Number: 18/471,661 Page 20 Art Unit: 2123
