CTFR 17/720,247 CTFR 79409 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. Claims 1-24 are presented for the examination. Claim Rejections - 35 USC § 103 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, 7, 13, 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) . As to claim 1 , Katzenberger teaches circuitry to, in response to an application programming interface (API) call cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified( The SoC may include an integrated circuit chip that includes one or more of a processor (e.g., a central processing unit (CPU), microcontroller, microprocessor, digital signal processor (DSP), etc.), memory, one or more communication interfaces, and/or further circuits, and may optionally execute received program code and/or include embedded firmware to perform functions, para[0059], ln 17-25/ When a pipeline is submitted for execution (e.g., during execution of application 126 ), the API calls to operators 112 via API 110 may be monitored by API detector 116 . For each API call for an operator of operators 112 that is detected, API detector 116 provides an indication 132 to graph builder 122 . Indication 132 may specify at least one of a name of the corresponding operator, one or more inputs for the corresponding operator, one or more outputs for corresponding operator, another node from which the one or more inputs are provided and/or another node to which the one or more outputs are provided. For each indication 132 received, graph builder 122 generates a node corresponding to the operator of operators 112 for an execution graph 118 . Execution graph 118 is representative of the machine learning pipeline developed by the developer. Each node in execution graph 118 may specify the information included in indicator 120 . For example, each node may specify at least one of a name of the corresponding operator, one or more inputs for the corresponding operator, one or more outputs for corresponding operator, another node from which the one or more inputs are provided and/or another node to which the one or more outputs are provided. The nodes and execution graph 118 may formatted in accordance with a JavaScript Object Notation (JSON) file and may be structured in accordance with the programming language in which application 126 is structured (e.g., Python). Graph builder 122 continues to update execution graph 118 with nodes until API detector 116 detects an API call for a function that is configured to train the machine learning model …….. As described above, graph execution engine 120 may be structured in accordance with a second programming language (e.g., C#). Graph execution engine 120 reconstructs (or generates) a second execution graph (e.g., execution graph 128 ) based on execution graph 118 . Execution graph 128 is structured in accordance with the programming language in which graph execution engine 120 is structured. For instance, graph execution engine 120 analyzes the nodes, input, outputs, etc., specified in first execution graph 120 and generates nodes, inputs, outputs, etc. that are in accordance with the second programming language, para[0033] to para[0034] to para[0035], ln 1-3/ para[0036], ln 1-12), Curzi teaches cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified( discussed above with respect to FIG. 2, the topology of the webpage 201 may be determined by traversing the tree-like structure[ executable graph code] of the DOM of the webpage 201, col 27, ln 19-22/ The DOM provides an application programming interface (API) that defines a logical structure of an HTML or XML document and provides functions for accessing and manipulating the content of the document. The DOM [ API ]provides functions for adding, modifying, and deleting elements from the DOM , col 5, ln 60-67/ The WAF module 135 may analyze the structure of the DOM of the webpage 201 using various functions of by an Application Programming Interface (API) of the DOM to traverse the structure of the webpage 201 . The shadow graph 209 represents a subset of the structure of the DOM of the webpage 201 . The WAF 135 may identify editable nodes in which a user may add, modify, and/or delete textual content while traversing the DOM of the webpage 201 and a representation of these editable nodes as a node in the shadow graph 209 , col 6, ln 30- 42/ The WAF module 135 extracts the contents of these editable nodes using functions provided by the API of the DOM of the webpage 201 and creates a node or nodes in the shadow graph 209 that represent these editable nodes. The nodes of the shadow graph may be linked back to the corresponding DOM objects of the webpage 201 , and the text contents of the corresponding DOM objects may be monitored in order to keep the contents of the shadow graph up to date , col 6, ln 55-45/ eferring again to FIG. 2, the shadow graph 209 is configured to monitor the DOM of the webpage 201 for changes that indicate that the structure and/or contents of the shadow graph 209 may need to be updated to keep the shadow graph 209 synchronized with the webpage 201 . The shadow graph 209 may be configured to listen for mutation events generated by the DOM as a result of modifications being made to the structure and/or the contents of the DOM, col 9, ln 55-65). It would have been obvious to one of the ordinary skill in the art before the effective filing date of claimed invention was made to modify the teaching of Katzenberger with Curzi to incorporate the feature because this allows the model of the webpage is configured to automatically monitor the topology of the webpage and to update the model based on detected changes to the webpage. As to claims 7, 13, 19, they are rejected for the same reason as to claim 1 above. In addition, Hsu teaches processor, non-transitory machine-readable medium(processor, col 28, ln 47-50/ A non-transitory computer readable storage, claim 17, ln 1-2) . 07-21-aia AIA Claim (s) 3, 8, 10, 14, 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1)in view of Russell( US 10949178 B1) . As to claim 3, Russell teaches wherein the one or more circuits are to modify before the automatically-generated graph code is completely generated, the one or more portions based, at least in part, on an identifier corresponding to the executable graph code (In a graph projection of an API, nodes in the graph projection of the API may represent operations that may be performed on a specific object, and connections between different objects may represent relationships where, col 3, ln 48-53/ To truncate[ generate ] the subgraph projection of the API, API subgraph generator 134 can identify, for each node in the subgraph projection, whether a node representing a type reference is included in the subgraph projection. If a node representing the type reference is not included in the subgraph projection, the type reference may be replaced in the subgraph projection with a leaf type, which may be a reference to a type that does not include any further type references. By replacing type references with leaf types, the subgraph projection of the API may exclude large graphs of types or other nodes that are not needed by the application 122 , col 8, ln 60-67/ in order to truncate [ generate ] the subgraph, the portion of graph code is modified, therefore, the portion of graph code is modified before the automatically-generated graph code is completely generated as described above./ FIG. 3 illustrates example operations 300 for generating[ generate] a subgraph of an application programming interface (API) based on information about queries invoked by an application…. , col 10, ln 9-15/ At block 330 , the system generates a subgraph projection of the API including the identified set of nodes. Generally, the subgraph projection of the API may be generated as a minimum spanning tree including at least some of the identified set of nodes. The nodes included in the subgraph projection of the API may include nodes identified a priori as types that are most often included in a path between any pair of types in the graph projection of the API and other nodes that represent origin nodes and nodes discovered using a depth-first traversal from the a priori identified nodes (if any) included in the subgraph projection. In some embodiments, the subgraph projection may be truncated to remove certain nodes that represent missing objects in the subgraph by replacing these nodes with terminating nodes that do not include references to other objects or data types. These missing objects generally include objects that may be referenced recursively by another object in the subgraph but is not included in the subgraph. By removing nodes representing missing objects and replacing nodes representing missing objects with terminating nodes, recursive references may be eliminated, which may exclude portions of the full graph that are not needed by an application from being included in the generated subgraph, col 11, ln 9-31/ Fig. 3 for generating a subgraph of an application programming interface (API) which includes step 330 described how to modify the portion of graph in order to completely generate the subgraph api therefore the modifying before the generation of subgraph api ) . It would have been obvious to one of the ordinary skill in the art before the effective filing date of claimed invention was made to modify the teaching of Katzenberger and Curzi with Russell to incorporate the feature because this generates and deploys code implementing an API that reduces the use processing and storage resources. As to claim 8, Russell teaches wherein the one or more processors are to modify, during generation of a graph associated with the executable graph code , the one or more portions by at least causing a first graph code node to depend on a second graph code node of the one or more nodes ( FIG. 4 illustrates further details regarding operations 330 (illustrated in FIG. 3) for generating the subgraph projection of the API including the identified set of nodes. As illustrated, operations 330 begin at block 410 , where a system builds a subgraph projection of the API as a spanning tree from the identified set of nodes. The subgraph projection of the API may be built such that the identified set of nodes are connected using a minimal number of edges. Generally, the identified set of nodes may include nodes representing objects referenced in application source code and nodes representing objects that are known a priori to be included in a set of paths between pairs of objects in the graph projection of the API. At block 420 , the system traverses the graph projection of the API to add additional nodes to the subgraph according to a maximum traversal depth defined for nodes in the subgraph. Generally, traversing the graph projection of the API according to a maximum traversal depth for nodes in the subgraph allows for indirect references to objects in the subgraph to be resolved and added to the subgraph projection of the API. The maximum traversal depth may be defined on a per-node (i.e., per-object or per-data-type) basis such that different nodes may have different maximum traversal depths or on a global basis such that each node has the same maximum traversal depth for adding additional nodes to the subgraph. In traversing the graph projection of the API, the node in the graph projection of the API being analyzed may be designated as the root node of a search. A breadth-first search may be performed from the node to identify nodes to include in the subgraph up to the maximum traversal depth below the node. At block 430 , the system truncates the subgraph projection of the API to remove references to objects not included in the subgraph projection of the API. To truncate the subgraph projection of the API, the system can examine references to other objects or data types included in each node of the subgraph projection of the API. If a node representing an object references another object that is not represented by a node in the subgraph, the referenced object may be removed from the subgraph and replaced with a terminating node. The terminating node may be a node that does not include any references to other objects in the subgraph so that unneeded objects are not included in the subgraph projection of the API. At block 440 , the system outputs the generated subgraph projection of the API. As discussed, the generated subgraph projection of the API may be deployed to an application server for use with an application, compiled into one or more libraries deployed with an application executable, or otherwise used in conjunction with an application in lieu of a larger, global API that includes queries that are not invoked by the application, col 12, ln 46-67 / col 13, ln 1-32) for the same reason as to claim 3 above. As to claim 10 , Russell teaches the incomplete graph code indicates one or more operations and dependencies among the one or more operations( col 8, ln 1-10) for the same reason as to claim 3 above. As to claim 14 , Russell teaches the set of instructions further include instructions, which if performed by the one or more processors, cause the one or more processors to modify, while at least a portion of the graph associated with the one or more portions of executable graph code are being built, a set of dependencies associated with the one or more portions( At block 330 , the system generates a subgraph projection of the API including the identified set of nodes. Generally, the subgraph projection of the API may be generated as a minimum spanning tree including at least some of the identified set of nodes. The nodes included in the subgraph projection of the API may include nodes identified a priori as types that are most often included in a path between any pair of types in the graph projection of the API and other nodes that represent origin nodes and nodes discovered using a depth-first traversal from the a priori identified nodes (if any) included in the subgraph projection. In some embodiments, the subgraph projection may be truncated to remove certain nodes that represent missing objects in the subgraph by replacing these nodes with terminating nodes that do not include references to other objects or data types, col 11, ln 9-30) for the same reason as to claim 3 above. As to claim 20 , It is rejected for the same reason as to claim 14 above . 07-21-aia AIA Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of Breeds(US 20100079462 A1) . As to claim 2 , Breeds teaches modify the one or more portions by at least modifying one or more dependencies associated with the incomplete graph code ( The server graph creation function 18 performs outbound and inbound impact analysis calls to the API of an impact analysis module 14A as required, para[0079]/ the server graph creation function 18 converts each result item to a graph edge object, adding to a graph edge list, and a graph node. All objects and edges are given a unique ID, para[0081]/The size of the viewing window 62 is fixed and represents the contents of the entire contents (objects) graph structure 63. If the graphical view window 42 is sufficiently large to display all objects in the graph structure 63, the view port control 64 extends to the same size as the viewing window 62. Similarly, user movement of the scroll bars 61 on the graphical view window 62 will change the visible graph contents and this is reflected in the viewing window 62, with the view port rectangle 64 automatically moved accordingly, para[0066], ln 3-16/ From the caller's request for a certain level of reliability, the API selects an error correction strategy and inserts encoder/decoder nodes 89 and 90 into the graph data structure 12, col 11, ln 29-33). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with Breeds to incorporate the feature of modify the one or more portions by at least modifying one or more dependencies associated with the incomplete graph code because this generates a graph view of multiply connected objects in a computing environment . 07-21-aia AIA Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of McAuley(US 20110082670 A1) . As to claim 4, McAuley teaches the incomplete graph code encodes one or more operations executable by one or more graphics processing units (GPUs)( a different digital device, such as a graphical processing unit (GPU), network server, or so forth embody these systems. Further, it is contemplated for the training system of FIG. 3 and the object comparator of FIG. 4 to be embodied by different digital devices. For example, the training system of FIG. 3, para[0032], ln 1-10/ The operation 26 suitably employs the processing described herein with reference to FIG. 1 in which the vertex model graph =(V,E.sup.VM) of low maximal clique size is generated, and the processing described herein with reference to FIG. 2 in which the vertex model graph is extended to include missing edges (if any) by addition of replicate nodes and the missing edges to form the globally rigid model graph =(V.sup.ext,E.sup.M) of low maximal clique size that encodes all edges of the original (first) graph, para[0037], ln 6-18). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with McAuley to incorporate the feature of the incomplete graph code encodes one or more operations executable by one or more graphics processing units (GPUs) because this is optimizing a mapping between vertices of the first model graph and vertices of the second graph . 07-21-aia AIA Claim (s) 5 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of Desai( US 20200312005 A1) As to claim 5 , Desai teaches wherein the one or more dependencies of the software graph indicate one or more constraints on scheduling an accelerator to perform a one or more operations corresponding to the one or more nodes( based on the particular operation, determines a graph data structure that includes a plurality of graph nodes, each of the plurality of graph nodes defining a set of constraints for performing a respective one of the set of tasks. In such an embodiment, the evaluating of the set of tasks includes analyzing the graph data structure to determine a distribution plan for the distributing. In some embodiments, one of the plurality of graph nodes specifies a constraint (e.g., desired compute capabilities 422 ) for using particular hardware to perform one of the set of tasks, and the evaluating includes identifying a compute node having the particular hardware for performing the task. In some embodiment, the particular hardware is a graphics processing unit (GPU), para[0092], ln 3-1-18). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with Desai to incorporate the feature of teach API is a runtime API because this identified abilities of the one or more compute nodes to facilitate the rendering . 07-21-aia AIA Claim (s) 6, 9, 11, 15, 22, 23 are rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of Manion( US 20040111469 A1) . As to claim 6, Manion teaches the one or more circuits are further to obtain a status indication as a result of performing the API( First, the peer graph add record API is used to add a new record to the graph as introduced above. A record added with this API is flooded to each node in the graph. The parameters for this API include the graph handle, a pointer to record data, and a pointer that is set to the record ID that uniquely identifies a record in a graph. It is noted that only the size, type, and expiration are required in the record, while the data and attributes are optional. This API returns an indication of success or failure, para[0082], ln 5-16/ To initiate shut down, the application calls the Peer Graph Shutdown function. This function cleans up any resources allocated by the call to the API Peer Graph Startup. As indicated above, there is preferably one call to Peer Graph Shutdown for each call to Peer Graph Startup. There are no required parameters for this function. This function returns an indication of success or failure of the shutdown function. In the case of error, the function returns an appropriate error code., para[0064]). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with Manion to incorporate the feature of the one or more circuits are further to obtain a status indication as a result of performing the API because this provides an indication as to the problem resulting in the failure. As to claim 9, Manion teaches the one or more processors are to cause the one or more portion of the executable graph code to be modified based at least in part on one or more parameter values indicating a set of graph code nodes(para[0067]/ para[0080, ln 1-20) for the same reason as to claim 6 above . As to claim 11, Manion teaches the one or more processors are to cause the one or more portion of the executable graph code to be modified based at least in part on one or more parameter values indicating one or more data types( para[0080, ln 1-20) for the same reason as to claim 6 above . As to claim 15 , Manion teaches cause the one or more portion of the executable graph code to be modified based at least in part on one or more flags indicated by one or more parameter values associated with the API( para[0067]/para[0080, ln 1-20) for the same reason as to claim 6 above. As to claim 22 , Manion teaches the incomplete graph code is graph code being generated through one or more processes( para[0106]/ para[0082]/ para[0083]) for the same reason as to claim 6 above. As to claim 23 , Manion teaches comprising obtaining one or more status indications indicating results of performing the API( para[0082], ln 5-16) for the same reason as to claim 6 above . 07-21-aia AIA Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of VENKATARAMA(US 20140281727 A1) . As to claim 12 , Venkatarama teaches the incomplete graph code indicates a set of operations executable by one or more general purpose graphics processing units (GPGPUs)( For instance, a general purpose GPU (GPGPU) programming environment may include thousands of GPGPUs, each running tens of thousands of threads, processing the same code in order to reach a result, such as, rendering a graphical image, para[0005], ln 614/ All the connected components in the graph are related to hazards that involve two different accesses locations. For example, a pair of vertices in the graph could be related to hazards involving multiple read operations and a single write operation with an incorrect address, para[0076], ln 5-14). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Hsu withVenkatarama to incorporate the feature of the incomplete graph code indicates a set of operations executable by one or more general purpose graphics processing units (GPGPUs) because this allows traversing the graph to report an analysis record for each hazard represented in the graph . 07-21-aia AIA Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of BOO(US 20180349113 A1) . As to claim 16 , BOO teaches the API is a driver API(the processes shown in FIGS. 4-5 can be performed by a sub-component of the driver called the “translator provider.” The translator provider can be a software function or hardware implemented logic which, given a job, para[0062], ln 1-5/ other aspects of the driver, the translator provider API is an extension point, para[0063], ln 14-16/ The translator provider component of the driver may automatically generate suitable translators, look up translators, or perform any other functionality which results in a translator being provided to translate the input job into a translation, para[0065], ln 5-12/ The Job API is mostly an extension point, intended for external implementation, para[0050], ln 7-9/ Returning to FIG. 1, at step 102 each job in the job graph is translated into a set of candidate translations corresponding to that job based at least in part on the one or more available engines. As discussed above, the one or more available engines can be received by the driver when the job graph is received, para[0052], ln 1-10). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with BOO to incorporate the feature of the API is a driver API because this improves the overall program performance as the source database can sometimes process transformation logic faster than the native environment of the program . 07-21-aia AIA Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1)and further in view of Belley( US 6469705 B1) . As to claim 17 , Belley teaches one or more portions indicate one or more memory allocation operations( The structure shown in FIG. 5 is known as a processing graph or dependency graph, and is used to determine allocation of system processing and memory resources in order to perform operations for a particular set of processing operations for a scene, each represented by a processing node in the processing graph, col 5, ln 50-55). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with Belley to incorporate the feature of each one or more portions indicate one or more memory allocation operations because this makes efficient use of the resources of the secondary cache and of the primary cache . 07-21-aia AIA Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) and further in view of Vadivelu(US 20190213067 A1) . As to claim 18 , Vadivelu teaches the incomplete graph code encodes operations executable by one or more central processing units (CPUs)( on remediation information associated with the issue in the graph database, para[0032], ln 11-14/ he encoding engine 140 may encode the present state of the module in a graph database, where the graph database comprises a set of representations of the module, para[0030], n 1-5/ Edge device 100 , which facilitates graph-based issue detection and remediation, may comprise a physical processor 110 , a representation engine 130 , an encoding engine 140 , an issue determination engine 150 , issue remediation engine 160 , and/or other engines, para[0028], ln 3-15/ The edge device (e.g., edge device 100 ) may comprise an access point, network switch, cloud server, or other hardware device that comprises a physical processor that implements machine readable instructions to perform functionality. The physical processor may be at least one central processing unit (CPU), microprocessor, and/or other hardware device suitable for performing the functionality described in relation to FIG. 2. In some examples, an edge device (e.g., edge device 100 ) may run a set of modules, para[0016], ln 1-20). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi with Vadivelu to incorporate the feature of teach the incomplete graph code encodes operations executable by one or more central processing units (CPUs) because this determines and/or remediates an issue after it has occurred . 07-21-aia AIA Claim 21 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) in view of Manion( US 20040111469 A1) and further in view of CHEN(CN 102467728 A) . As to claim 21 , Manion teaches based at least in part on a flag indicating ( First, the peer graph add record API is used to add a new record to the graph as introduced above. A record added with this API is flooded to each node in the graph. The parameters for this API include the graph handle, a pointer to record data, and a pointer that is set to the record ID that uniquely identifies a record in a graph. It is noted that only the size, type, and expiration are required in the record, while the data and attributes are optional. This API returns an indication of success or failure, para[0082], ln 5-16) for the same reason as to claim 6 . Chen teaches if there is the state directed graph network is not changed, then what does not need to do if the state is changed to "change", then searching all marked in the directed graph is "newly added" or "deleting" a directed edge, and marked as "modified" node, according to these nodes and directed edges. calculating the change and outputting in the directed graph, a whole directed graph status is "not change", each node in the graph, each directed edge mark as the unchanged, the output module. for the core engine module to generate the directed graph information into the transaction information to acquire trading scheme, and allocated to the corresponding user and article so as to finish reorganization of data, sec: claim 7, ln 66-89/ Referring to FIG 15, the modifying operation comprises: 1) adding one node, 2) deleting a node, 3) increasing one directed edge, calling this method returns after adding the side related to the side of ring number and half rings, 4) deleting one directed edge, 5) modifying node parameter, para[0107]). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger and Curzi and Manion with Chen to incorporate the feature of teach performing the API based at least in part on a flag indicating because this indicates that the edge is newly added or has been deleted, or with a change flag for each one node, a parameter indicative of the node has changed . 07-21-aia AIA Claim 24 is rejected under 35 U.S.C. 103 as being unpatentable over Katzenberger( US 20200348912 A1) in view of Curzi (US 11106757 B1) in view of McAuley(US 20110082670 A1) and further in view of Cosic( US 10474934 B1) . As to claim 24, McAuley teaches the incomplete graph code encodes one or more operations (a different digital device, such as a graphical processing unit (GPU), network server, or so forth embody these systems. Further, it is contemplated for the training system of FIG. 3 and the object comparator of FIG. 4 to be embodied by different digital devices. For example, the training system of FIG. 3, para[0032], ln 1-10/ The operation 26 suitably employs the processing described herein with reference to FIG. 1 in which the vertex model graph =(V,E.sup.VM) of low maximal clique size is generated, and the processing described herein with reference to FIG. 2 in which the vertex model graph is extended to include missing edges (if any) by addition of replicate nodes and the missing edges to form the globally rigid model graph =(V.sup.ext,E.sup.M) of low maximal clique size that encodes all edges of the original (first) graph, para[0037], ln 6-18) for the same reason as to claim 6 above. Cosic teaches encodes a set of operations executable by one or more parallel processing units (PPUs)( processor 11 can be provided in a graphics processing unit (GPU), visual processing unit (VPU), or other highly parallel processing unit or circuit such as, for example, nVidia GeForce line of GPUs, AMD Radeon line of GPUs, and/or others. Such GPUs or other highly parallel processing units, col 60, ln 25-30/ modifying one or more instruction sets for operating an application or an object of the application, the application running on the processor circuit, col 11, ln 35-40/ the device is included in a neuron, a node, a vertex, or an element of a data structure. The data structure may include a neural network, a graph, col 6, ln 45-50/ Node 852 may also include a function for transforming or manipulating any data, object, data structure, and/or other item. Examples of such transformation functions include mathematical functions (i.e. addition, subtraction, multiplication, division, sin, cos, log, derivative, integral, etc.), object manipulation functions (i.e. creating an object, modifying an object, deleting an object, appending objects, etc.), data structure manipulation functions (i.e. creating a data structure, modifying a data structure, deleting a data structure, creating a data field, modifying a data field, deleting a data field, etc.), col 95, ln 65-67 to col 96, ln 1-10). It would have been obvious to one of the ordinary skill in the art before the effective filling date of claimed invention was made to modify the teaching of Katzenberger , Curzi and McAuley with Cosic to incorporate the feature of a set of operations executable by one or more parallel processing units (PPUs) because this provides needed for computing enabled systems and/or devices to be less dependent on or fully independent from user input. Response to the argument : A. Applicant amendment filed on 02/02/2026 has been considered but they are not persuasive: Applicant argued in substance that : (1) “ One or more processors, comprising: circuitry to, in response to an application programming interface (API) call, cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified. Applicant respectfully submits that the combination of Hsu and Russell fails to teach or suggest, at least, circuitry to "cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified," in response to an API call, as recited in claim 1. The Office cites Russell as teaching this aspect of the claim. See Office Action at pp. 4-5. " B. Examiner respectfully disagreed with Applicant's remarks: As to the point (1), the amended claim feature recites “ in response to an application programming interface (API) call cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified” which indicated the modification at runtime. Katzenberger teaches “ in response to an application programming interface (API) call cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified” which indicated the modification at runtime ( When a pipeline is submitted for execution (e.g., during execution of application 126 ), the API calls to operators 112 via API 110 may be monitored by API detector 116 . For each API call for an operator of operators 112 that is detected, API detector 116 provides an indication 132 to graph builder 122 . Indication 132 may specify at least one of a name of the corresponding operator, one or more inputs for the corresponding operator, one or more outputs for corresponding operator, another node from which the one or more inputs are provided and/or another node to which the one or more outputs are provided. For each indication 132 received, graph builder 122 generates a node corresponding to the operator of operators 112 for an execution graph 118 . Execution graph 118 is representative of the machine learning pipeline developed by the developer. Each node in execution graph 118 may specify the information included in indicator 120 . For example, each node may specify at least one of a name of the corresponding operator, one or more inputs for the corresponding operator, one or more outputs for corresponding operator, another node from which the one or more inputs are provided and/or another node to which the one or more outputs are provided. The nodes and execution graph 118 may formatted in accordance with a JavaScript Object Notation (JSON) file and may be structured in accordance with the programming language in which application 126 is structured (e.g., Python). Graph builder 122 continues to update execution graph 118 with nodes until API detector 116 detects an API call for a function that is configured to train the machine learning model …….. As described above, graph execution engine 120 may be structured in accordance with a second programming language (e.g., C#). Graph execution engine 120 reconstructs (or generates) a second execution graph (e.g., execution graph 128 ) based on execution graph 118 . Execution graph 128 is structured in accordance with the programming language in which graph execution engine 120 is structured. For instance, graph execution engine 120 analyzes the nodes, input, outputs, etc., specified in first execution graph 120 and generates nodes, inputs, outputs, etc. that are in accordance with the second programming language, para[0033] to para[0034] to para[0035], ln 1-3/ para[0036], ln 1-12), Curzi teaches cause one or more portions of executable graph code to be modified to cause one or more nodes or one or more dependencies of a software graph corresponding to the executable graph code to be modified( discussed above with respect to FIG. 2, the topology of the webpage 201 may be determined by traversing the tree-like structure[ executable graph code] of the DOM of the webpage 201, col 27, ln 19-22/ The DOM provides an application programming interface (API) that defines a logical structure of an HTML or XML document and provides functions for accessing and manipulating the content of the document. The DOM [ API ]provides functions for adding, modifying, and deleting elements from the DOM , col 5, ln 60-67/ The WAF module 135 may analyze the structure of the DOM of the webpage 201 using various functions of by an Application Programming Interface (API) of the DOM to traverse the structure of the webpage 201 . The shadow graph 209 represents a subset of the structure of the DOM of the webpage 201 . The WAF 135 may identify editable nodes in which a user may add, modify, and/or delete textual content while traversing the DOM of the webpage 201 and a representation of these editable nodes as a node in the shadow graph 209 , col 6, ln 30-42/ The WAF module 135 extracts the contents of these editable nodes using functions provided by the API of the DOM of the webpage 201 and creates a node or nodes in the shadow graph 209 that represent these editable nodes. The nodes of the shadow graph may be linked back to the corresponding DOM objects of the webpage 201 , and the text contents of the corresponding DOM objects may be monitored in order to keep the contents of the shadow graph up to date , col 6, ln 55-45/ referring again to FIG. 2, the shadow graph 209 is configured to monitor the DOM of the webpage 201 for changes that indicate that the structure and/or contents of the shadow graph 209 may need to be updated to keep the shadow graph 209 synchronized with the webpage 201 . The shadow graph 209 may be configured to listen for mutation events generated by the DOM as a result of modifications being made to the structure and/or the contents of the DOM, col 9, ln 55-65). 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. Conclusion US 20220166679 A1 teaches In some embodiments, an agent comprises a collection of callback functions. For example, different functions may be executed based on whether a portion of a graph model associated with the triggering pattern was added to, modified in, or deleted from the graph model (e.g., whether a portion of the graph model is changed to match the triggering pattern, a property of an edge or node in a portion of the graph model that matches the triggering pattern is changed, or a portion of the graph model matching the triggering pattern is changed to no longer match the triggering pattern). The agent may store multiple functions, wherein the functions are executed based on a type of change in a portion of a graph model associated with the triggering pattern (e.g., “added,” “modified,” or “deleted”), a type of a changed data structure, a position of a changed data structure, a reference/path to a data structure, or any other factor. US 11086709 B1 teachesThe graph representation is updated based on any changes caused by agent callback functions. In some embodiments, the changes to the graph representation caused by the callback function invoke one or more additional callback functions. In some embodiments, the graph representation accurately represents the network configuration at any given time. Changes to the network configuration may be implemented by changing the graph representation, wherein changing the graph representation triggers agents to perform callback functions that execute the changes. US 20180210927 A1 teaches llback functions of invoked agents are invoked. In some embodiments, an agent is associated with a triggering pattern and a callback function. In the event a triggering pattern of an agent is detected, the agent is invoked and a callback function associated with the agent is invoked. The callback functions execute commands (e.g., to implement at least a portion of the intent). For example, the graph model is updated and network devices are configured by the callback functions triggered by detected changes to the appropriate portions of the graph representation associated with triggering patterns. In some embodiments, using a publish-subscribe model of triggering patterns and callback functions, changes to the network configuration are able to be implemented incrementally. US 11210143 B1 teaches The user device may then execute one or more calls to a runtime API of the workflow platform to update the workflow (e.g., updating the workflow graph and/or context of the workflow) maintained by the workflow platform. For example, the user device may spawn a first thread for executing the workflow update via the runtime API. The first thread may obtain. US 20120284792 A1 teaches Run-time: when the transformed program runs in memory. The new function-call structure self-modifies and changes the function-call graph. As the program executes the entire function-call graph is modified by modifying when and how functions are called. Any inquiry concerning this communication or earlier communications from the examiner should be directed to LECHI TRUONG whose telephone number is (571)272-3767. The examiner can normally be reached 10-8 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 Young Kevin can be reached on (571)270-3180. 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 571-272-10. /LECHI TRUONG/ Primary Examiner, Art Unit 2194 Application/Control Number: 17/720,247 Page 2 Art Unit: 2194 Application/Control Number: 17/720,247 Page 3 Art Unit: 2194 Application/Control Number: 17/720,247 Page 4 Art Unit: 2194 Application/Control Number: 17/720,247 Page 5 Art Unit: 2194 Application/Control Number: 17/720,247 Page 6 Art Unit: 2194 Application/Control Number: 17/720,247 Page 7 Art Unit: 2194 Application/Control Number: 17/720,247 Page 8 Art Unit: 2194 Application/Control Number: 17/720,247 Page 9 Art Unit: 2194 Application/Control Number: 17/720,247 Page 10 Art Unit: 2194 Application/Control Number: 17/720,247 Page 11 Art Unit: 2194 Application/Control Number: 17/720,247 Page 12 Art Unit: 2194 Application/Control Number: 17/720,247 Page 13 Art Unit: 2194 Application/Control Number: 17/720,247 Page 14 Art Unit: 2194 Application/Control Number: 17/720,247 Page 15 Art Unit: 2194 Application/Control Number: 17/720,247 Page 16 Art Unit: 2194 Application/Control Number: 17/720,247 Page 17 Art Unit: 2194 Application/Control Number: 17/720,247 Page 18 Art Unit: 2194 Application/Control Number: 17/720,247 Page 19 Art Unit: 2194 Application/Control Number: 17/720,247 Page 20 Art Unit: 2194 Application/Control Number: 17/720,247 Page 21 Art Unit: 2194 Application/Control Number: 17/720,247 Page 22 Art Unit: 2194 Application/Control Number: 17/720,247 Page 23 Art Unit: 2194 Application/Control Number: 17/720,247 Page 24 Art Unit: 2194
Prosecution Insights