Prosecution Insights
Last updated: July 17, 2026
Application No. 18/465,492

OPTIMIZING INSTRUCTION SCHEDULING AND MEMORY ALLOCATION FOR TENSOR AND GRAPHICAL PROCESSORS USING LATTICE IMAGE DATA STRUCTURE OPTIMIZATIONS

Final Rejection §102§103§112
Filed
Sep 12, 2023
Examiner
WANG, RONGFA PHILIP
Art Unit
2199
Tech Center
2100 — Computer Architecture & Software
Assignee
Groq Inc.
OA Round
2 (Final)
84%
Grant Probability
Favorable
3-4
OA Rounds
2m
Est. Remaining
91%
With Interview

Examiner Intelligence

Grants 84% — above average
84%
Career Allowance Rate
459 granted / 544 resolved
+29.4% vs TC avg
Moderate +7% lift
Without
With
+7.0%
Interview Lift
resolved cases with interview
Typical timeline
3y 0m
Avg Prosecution
12 currently pending
Career history
560
Total Applications
across all art units

Statute-Specific Performance

§101
3.2%
-36.8% vs TC avg
§103
66.2%
+26.2% vs TC avg
§102
17.8%
-22.2% vs TC avg
§112
10.5%
-29.5% vs TC avg
Black line = Tech Center average estimate • Based on career data from 544 resolved cases

Office Action

§102 §103 §112
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 . Detail Action This office action is in response to the amendment filed on 2/16/2026. Claims 1-18 are pending. Claim Rejections - 35 USC § 112 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. Claims 6-13 are rejected because claim 6 recites the limitation “said initial first program” in “lowering a plurality of operations in said initial first program …”. There is insufficient antecedent basis for this limitation in the claim. Claims 7-13 depend on claim 6 and inherit the limitation and deficiency and are rejected for similar reason. Claims 9 and 10 is/are rejected because claims 9/10 recites the limitation of “said step”. There is insufficient antecedent basis for this limitation in the claim. Claims 2-5 and 16-18 are rejected because claim 2 recites the limitation of “The compiler” There is insufficient antecedent basis for this limitation in the claim. Claims 3-5 and 16-18 depend on claim 2 and inherit the limitation and deficiency and are rejected for similar reason. Claims 16-18 are rejected because they recite the limitation of “the compiler process of claim 2” while claim 2 only recites the compiler. There is insufficient antecedent basis for this limitation in the claim. Claim 15 is rejected because it recites the limitation of “the hidden dimensions” and “the iteration spaces” There is insufficient antecedent basis for this limitation in the claim. Claim 18 is rejected because it recites the limitation of “the found lattice point” and “the found lattice point” There is insufficient antecedent basis for this limitation in the claim. 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. Claim(s) 1-5 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Ren et al. (US 2022/0413862 Al) Per claim 1, Ren discloses providing a lattice image data structure including a plurality of hidden dimensions for representing relations in a tensor program code, and executing an algorithm that combines the relations while reducing dimensionality of the lattice image data structure by reducing the plurality of hidden dimensions. ([0004], discloses operator fusion (kernel/layer fusion) of optimizing compiler. [0007-0008], discloses graph rewriting graph to generate and optimizing the fusion code. [0022-0023], disclose DNNFucion using graph corresponding to lattice and transformation of the graph to fuse operations to combine operations corresponding to kernels. When combining operations, the fused layers are reduced via applying rules on operations on tensor. [0058], see DNNFusion using graph rewriting to remove unnecessary operations…there are 18% fewer fused layers. [0005], it improves on Polyhedral analysis to captures combinations in DNN. See [0007]) Per claim 2, the rejection of claim 1 is incorporated. Ren discloses wherein the algorithm is a Kernel Characterization Algorithm KCA algorithm. ([0023], disclose algorithms for determine function of specific operations, where kernels correspond to operations and the algorithm is characterized by operations/kernels)) Per claim 3, the rejection of claim 2 is incorporated. Ren discloses wherein the KCA algorithm infers a low-dimensional shape of points from the hidden dimension set to reduce the hidden dimension of the resulting set or relation. ([0036], disclose mapping relationships for implementation of fusion optimization. TABLE 1, see reorganize, reshape… ) Per claim 4, the rejection of claim 2 is incorporated. Ren discloses wherein the KCA algorithm iteratively modifies the starting lattice basis until a desired property is achieved. ([0064], see the process is repeated until…) Per claim 5, the rejection of claim 2 is incorporated. Ren discloses wherein the KCA algorithm is run iteratively to validate a candidate shape or find missing points and to use missing points to improve the lattice basis. ([0099], see evaluating model…shape) 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. Claim(s) 6-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ren et al. (US 2022/0413862 Al) and further in view of Eberle et al. (US 2015/0193233 A1) Per claim 14, Ren discloses An apparatus enabling a compiler in interconnected accelerator units to simplify a machine learning (ML) graph representing a program to be compiled, said apparatus including a processor to: ([0023], see graph rewriting applying rules on operations of tensors to fuse operations.) lower a plurality of operations in an initial first program to an original plurality of at least two loops; ( [0022], see loop fusing via classifying operators into different types and combination. The combination lower operations. [0061], graph rewriting to reduce size) and infer a fused loop structure from said original plurality of at least two loops in said initial first program thus creating a second program having said fused loop, ([0058], optimize ECG to remove unnecessary operations….operations that are common….18% fused layers ) Ren does not, however, Eberle discloses wherein said fused loop in said second program is pipelining multiplication and addition operations. ([0119], see overlapping / interleaving multiplication operation and addition operation.) Therefore, it would have been obvious to a person of ordinary skill before the effective filing date of the invention to incorporate the teachings of Eberle into the teachings of Ren to include the limitation disclosed by Eberle. The modification would be obvious to one of ordinary skill in the art to want to improve program performance as suggested by Eberle ([0119]) Per claim 6, it a method claim corresponding to apparatus claim 14 and is rejected for similar reasons above. Per claim 7, the rejection of claim 6 is incorporated. Ren/Eberle discloses wherein said interconnected accelerator units are a tensor streaming processor (TSP). (Ren, [0087], see various types of processors to perform tensor processing) Per claim 8, the rejection of claim 6 is incorporated. Ren/Eberle discloses expressing indexing logic in said original plurality of loops in said initial first program as lattice relations. (Ren, [0076], see index mapping …) Per claim 9, the rejection of claim 7 is incorporated. Ren/Eberle discloses calculating the hidden dimensions of the bijection between the iteration spaces of at least two initial loops. (Ren, [0067], see one-to-one mapping corresponding to bijection) Per claim 10, the rejection method of claim 8 is incorporated. Ren/Eberle discloses said step further comprising applying Kernel Characterization Algorithm (KCA) to said lattice relations. (Ren, [0023], disclose algorithms for determine function of specific operations, where kernels correspond to operations) Per claim 11, the rejection method of claim 10 is incorporated. Ren/Eberle discloses wherein said KCA algorithm infers a low-dimensional shape of points from said hidden dimension set, thus reducing the hidden dimension of the resulting set or relation. (. Ren,[0022-0023], disclose DNNFucion using graph corresponding to lattice and transformation of the graph to fuse operations to combine operations corresponding to kernels. When combining operations, the fused layers are reduced via applying rules on operations on tensor. [0058], see DNNFusion using graph rewriting to remove unnecessary operations…there are 18% fewer fused layers) Per claim 12, the rejection method of claim 10 is incorporated. Ren/Eberle discloses wherein said KCA algorithm comprises the following steps: validating a candidate lattice basis; and using the missing points to improve said candidate lattice basis. ([0103], see enabling larger set of optimization) Per claim 13, the rejection method of claim 12 is incorporated. Ren/Eberle discloses further comprises: using Integer Linear Programming to validate said candidate lattice basis. (Ren, [0104], see DNNFusion …linear algebra computation corresponds to ILP) Per claim 15, the rejection of claim 14 is incorporated. Ren/Eberle discloses a Kernel Characterization Algorithm (KCA) configured to calculate the hidden dimensions of the bijection between the iteration spaces of at least two initial loops. (Ren, [0067], see one-to-one mapping corresponding to bijection) Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action. Allowable Subject Matter Claim 18 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 and overcoming issues identified in this office action. Conclusion The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Any inquiry concerning this communication or earlier communications from the examiner should be directed to Philip Wang whose telephone number is 571-272-5934. The examiner can normally be reached on Monday – Friday 8:00AM -4:00PM. Any inquiry of general nature or relating to the status of this application should be directed to the TC2100 Group receptionist: 571-272-2100. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Lewis Bullock, can be reached at 571-272-3759. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. 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). /PHILIP WANG/Primary Examiner, Art Unit 2199
Read full office action

Prosecution Timeline

Sep 12, 2023
Application Filed
Jan 09, 2024
Response after Non-Final Action
Sep 15, 2025
Non-Final Rejection mailed — §102, §103, §112
Feb 16, 2026
Response Filed
Jun 11, 2026
Final Rejection mailed — §102, §103, §112 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12681715
SYSTEM, CENTER, PROCESSING METHOD, AND STORAGE MEDIUM
2y 6m to grant Granted Jul 14, 2026
Patent 12669991
RECOMMENDING USER FEEDBACK WITH AUTOMATIC SCORING AND REASONING
2y 7m to grant Granted Jun 30, 2026
Patent 12650821
MACHINE LEARNING REPOSITORY SERVICE
2y 3m to grant Granted Jun 09, 2026
Patent 12650857
PROGRAM CALL STACK CREATION METHOD AND UNWIND STACK METHOD AND APPARATUS
2y 3m to grant Granted Jun 09, 2026
Patent 12650834
CONTINUOUS DEPLOYMENT OF MICROSERVICES OR OTHER APPLICATIONS USING SEPARATE SOURCE AND DEPLOYMENT REPOSITORIES
2y 1m to grant Granted Jun 09, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
84%
Grant Probability
91%
With Interview (+7.0%)
3y 0m (~2m remaining)
Median Time to Grant
Moderate
PTA Risk
Based on 544 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month