DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copy has been filed in parent Application No. JP2024-098009, filed on 6/18/2024.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 2/13/2025 is/are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is/are being considered by the examiner.
Allowable Subject Matter
Claims 2-5 and 13-15 is/are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter:
The most relevant prior art has been cited in the included form PTO-892 Notice of References Cited:
Regarding dependent claim 2,
SANCHEZ et al. (US PGPUB No. 2024/0111488; Pub. Date: Apr. 4, 2024)
SANCHEZ is directed to a data processing system configured to determine variances associated with variables of a dataset by processing said dataset using a trained model. FIG. 5 illustrates a process of determining causal relations between variables comprising multiple stages including stage 60 of determining causal relations using variables represented by leaf node order (See FIG. 5 & Paragraph [0042]-[0043]).
HORIWAKI et al. (US PGPUB No. 2018/0307219; Pub. Date: Oct. 25, 2018)
HORIWAKI is directed to a system in which a causal relation model is efficiently used and verified according to domain knowledge. The process comprises constructing a causal relation model using aggregated data. A subset and domain knowledge verification section 35 verifies a subset of monitor data using subset extraction section 34 and domain knowledge storage section 47, wherein domain knowledge may be applied to a causal relation model in order to resolve contrarieties between the causal relation model and knowledge domain data (See Paragraph [0051]).
CAO (China Invention Application Publication No.: CN 118018429 A; Pub. Date: May 10, 2024)
CAO is directed to a system for estimating a node topological order to optimize network topological efficiency. The method comprises steps S101-S105 wherein step S104 comprises performing Lasso estimation on non-zero elements of a weighted adjacent matrix of a Bayesian network model according to a second data matrix to obtain element weight values of the weighted adjacent metrics. At step S105, the system constructs a directed acyclic graph (DAG) of network service satisfaction metrics according to the weighted adjacency metrics (See Pg. 3, Paragraphs 4-13).
SUN et al. (US PGPUB No. 2022/0398260; Pub. Date: Dec. 15, 2022)
SUN is directed to an information processing method for determining independence relationships from among a plurality of variables. FIG. 1 illustrates the method comprising step S13 of adjusting a first independent relationship according to an adjustment scheme to obtain a second independence relationship. The system may adjust a first independent value so that the correct causal relationship information may be obtained as in step S14. Note [0074] wherein the system may reduce the impact of data noise using a composite score to determine the possibility that an independence relationship is correct (See FIG. 1 & Paragraph [0116]).
While the references identified above disclosed at least some of the limitations of independent claim 1, the references, neither alone nor in combination, do not disclose at least the following limitations:
wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, and the one or more hardware processors are configured to: calculate an exogenous noise matrix by multiplying a record data matrix including the one or more pieces of record data by a matrix obtained by subtracting the adjacency matrix from an identity matrix, the exogenous noise matrix including the plurality of exogenous noise estimation values for each of the one or more pieces of record data;
and determine whether a causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from the causal relationship of the plurality of variables that is represented by the pre-update model, based on the exogenous noise matrix.
While the references disclose the use of causal models to determine relationships between variables and some incorporate calculating noise metrics, the references neither alone nor in combination disclose the specific manner in which the exogenous noise matrix is calculated in claim 2, which requires a specific mathematical relationship involving a record data matrix, adjacency matrix and identify matrix.
Therefore, for at least the reasons above, claim 2 is allowable.
Regarding dependent claim 3,
Claim 3 is dependent upon claim 2 and therefore contains the same allowable subject matter and is objected to for at least the same reasons indicated above.
Regarding dependent claim 4,
Claim 4 is dependent upon claim 2 and therefore contains the same allowable subject matter and is objected to for at least the same reasons indicated above.
Regarding dependent claim 5,
Claim 5 is dependent upon claim 2 and therefore contains the same allowable subject matter and is objected to for at least the same reasons indicated above.
Regarding dependent claim 13,
The prior art discussed above with regard to dependent claim 2 is also the most relevant prior art to dependent claim 13.
While the references identified above disclosed at least some of the limitations of independent claim 1, the references, neither alone nor in combination, do not disclose at least the following limitations:
wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, and the one or more hardware processors are configured to: generate a plurality of first exogenous noise estimation values that are the plurality of exogenous noise estimation values for each of the one or more pieces of record data, based on the one or more pieces of record data and the pre-update model;
execute first determination processing of determining whether a causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from the causal relationship of the plurality of variables that is represented by the pre-update model, based on independence between any two or more variables in the plurality of first exogenous noise estimation values for each of the one or more pieces of record data;
output at least one piece of information from information indicating there is no change in causal influence, there is no change in causal structure, and there is no change in causal order when determining that the causal relationships are not different by the first determination processing, generate a first structural causal model that is the structural causal model by estimating a new value corresponding to a nonzero element in the adjacency matrix of the pre-update model using linear regression based on the one or more pieces of record data and updating a value of the nonzero element in the adjacency matrix of the pre- update model to the estimated new value when determining that the causal relationships are different by the first determination processing, generate a plurality of second exogenous noise estimation values that are the plurality of exogenous noise estimation values for each of the one or more pieces of record data, based on the one or more pieces of record data and the first structural causal model, execute second determination processing of determining whether the causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from a causal relationship of the plurality of variables that is represented by the first structural causal model, based on independence between any two or more variables in the plurality of second exogenous noise estimation values for each of the one or more pieces of record data;
and output at least one piece of information from information indicating there is a change in causal influence, there is no change in causal structure, and there is no change in causal order, when determining that the causal relationships are not different by the second determination processing.
While the references disclose the use of causal models to determine relationships between variables and some incorporate calculating noise metrics, the references neither alone nor in combination disclose the specific manner in which the system determines changes in causal relationships across a plurality of variables.
Therefore, for at least the reasons above, claim 13 is allowable.
Regarding dependent claim 14,
Claim 14 is dependent upon claim 13 and therefore contains the same allowable subject matter and is objected to for at least the same reasons indicated above.
Regarding dependent claim 15,
The prior art discussed above with regard to dependent claim 2 is also the most relevant prior art to dependent claim 13.
While the references identified above disclosed at least some of the limitations of independent claim 1, the references, neither alone nor in combination, do not disclose at least the following limitations:
wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, and the one or more hardware processors are configured to: generate a plurality of first exogenous noise estimation values that are the plurality of exogenous noise estimation values for each of the one or more pieces of record data, based on the one or more pieces of record data and the pre-update model;
execute first determination processing of determining whether the causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from the causal relationship of the plurality of variables that is represented by the pre-update model, based on independence between any two or more variables in the plurality of first exogenous noise estimation values for each of the one or more pieces of record data;
output at least one piece of information from information indicating there is no causal influence, there is no change in causal structure, and there is no change in causal order, when determining that the causal relationships are not different by the first determination processing;
generate a second structural causal model that is the structural causal model using regularized linear regression, based on the one or more pieces of record data while causing a causal order to be the same as the adjacency matrix of the pre-update model, when determining that the causal relationships are different by the first determination processing;
generate a plurality of third exogenous noise estimation values that are the plurality of exogenous noise estimation values for each of the one or more pieces of record data, based on the one or more pieces of record data and the second structural causal model; execute third determination processing of determining whether the causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from a causal relationship of the plurality of variables that is represented by the second structural causal model, based on independence between any two or more variables in the plurality of third exogenous noise estimation values for each of the one or more pieces of record data;
output at least one piece of information from information indicating there is a change in causal influence, there is a change in causal structure, and there is no change in causal order, when determining that the causal relationships are not different by the third determination processing;
and output at least one piece of information from information indicating there is a change in causal influence, there is a change in causal structure, and there is a change in causal order, when determining that the causal relationships are different by the third determination processing.
While the references disclose the use of causal models to determine relationships between variables and some incorporate calculating noise metrics, the references neither alone nor in combination disclose the specific manner in which the system determines changes in causal relationships across a plurality of variables.
Therefore, for at least the reasons above, claim 15 is allowable.
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) 1, 9-10, 17 and 20-23 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ et al. (US PGPUB No. 2024/0111488; Pub. Date: Apr. 4, 2024) in view of HORIWAKI et al. (US PGPUB No. 2018/0307219; Pub. Date: Oct. 25, 2018).
Regarding independent claim 1,
SANCHEZ discloses an information processing device comprising one or more hardware processors configured to: generate a plurality of exogenous noise estimation values corresponding to a plurality of variables for each of one or more pieces of record data including a plurality of record values respectively corresponding to the plurality of variables, based on the one or more pieces of record data and a pre-update model that is a structural causal model representing a causal relationship of the plurality of variables, See FIG. 5 & Paragraph [0042]-[0043], (Disclosing a data processing system configured to determine variances associated with variables of a dataset by processing said dataset using a trained model. FIG. 5 illustrates a process of determining causal relations between variables comprising multiple stages including stage 60 of determining causal relations using variables represented by leaf node order. Note [0033] wherein causal relations are considered present when one variable in a dataset has a direct influence on another variable.) See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) consisting of a collection of assignments of elements of a d-dimensional random vector "x" wherein each element xi of the vector is determined according to a mathematical relationship between parent nodes of xi in the DAG and a noise term independent of xi referred to as exogenous noise, i.e. generate a plurality of exogenous noise estimation values corresponding to a plurality of variables for each of one or more pieces of record data including a plurality of record values respectively corresponding to the plurality of variables, based on the one or more pieces of record data and a pre-update model that is a structural causal model representing a causal relationship of the plurality of variables,
the plurality of exogenous noise estimation values each representing estimation values of influence by exogenous noises that are different from influences from the plurality of variables with respect to corresponding variables among the plurality of variables; See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) consisting of a collection of assignments of elements of a d-dimensional random vector "x" wherein each element xi of the vector is determined according to a mathematical relationship between parent nodes of xi in the DAG and a noise term independent of xi referred to as exogenous noise.)
SANCHEZ does not disclose the step wherein the device may determine whether a causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from the causal relationship of the plurality of variables that is represented by the pre-update model, based on independence between any two or more variables in the plurality of exogenous noise estimation values with respect to each of the one or more pieces of record data;
and generate a post-update model that is the structural causal model based on the one or more pieces of record data when determining that the causal relationships are different.
HORIWAKI disclose the step wherein the device may determine whether a causal relationship of the plurality of variables that is represented by the one or more pieces of record data is different from the causal relationship of the plurality of variables that is represented by the pre-update model, based on independence between any two or more variables in the plurality of exogenous noise estimation values with respect to each of the one or more pieces of record data; See Paragraph [0051], (Disclosing a system in which a causal relation model is efficiently used and verified according to domain knowledge. The process comprises constructing a causal relation model using aggregated data. A subset and domain knowledge verification section 35 verifies a subset of monitor data using subset extraction section 34 and domain knowledge storage section 47, wherein domain knowledge may be applied to a causal relation model in order to resolve contrarieties between the causal relation model and knowledge domain data. Note [0036] wherein the corrected causal relation model may remove a contradictory link determined based on domain knowledge or add a link based on domain knowledge.)
and generate a post-update model that is the structural causal model based on the one or more pieces of record data when determining that the causal relationships are different. See Paragraph [0051], (Subset and domain knowledge verification section 35 verifies a subset of monitor data using subset extraction section 34 and domain knowledge storage section 47, wherein domain knowledge may be applied to a causal relation model in order to resolve contrarieties between the causal relation model and knowledge domain data.) See Paragraph [0036], (The corrected causal relation model may removes a contradictory link determined based on domain knowledge or add a link based on domain knowledge, i.e. generate a post-update model that is the structural causal model based on the one or more pieces of record data when determining that the causal relationships are different.)
SANCHEZ and HORIWAKI are analogous art because they are in the same field of endeavor, causal relation modelling. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ to include the method of updating a causal relationship model according to domain knowledge indicating relationships between elements of a dataset as disclosed by HORIWAKI. Paragraph [0064] of HORIWAKI discloses that the use of domain knowledge allows the system to specify the set of monitor data which is not contradictory to the domain knowledge and has a large index. This results in a reduction of the time required to verify the subset of the monitor data, which as influence on quality data, domain knowledge, etc.
Regarding dependent claim 10,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
HORIWAKI further discloses the step wherein the one or more hardware processors are configured to generate the post-update model including a causal order, based on the one or more pieces of record data when determining that the causal relationships are different. See Paragraph [0051], (Subset and domain knowledge verification section 35 verifies a subset of monitor data using subset extraction section 34 and domain knowledge storage section 47, wherein domain knowledge may be applied to a causal relation model in order to resolve contrarieties between the causal relation model and knowledge domain data.) See Paragraph [0036], (The corrected causal relation model may removes a contradictory link determined based on domain knowledge or add a link based on domain knowledge, i.e. generate the post-update model including a causal order, based on the one or more pieces of record data when determining that the causal relationships are different (e.g. by correcting a contradiction).)
Regarding dependent claim 17,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses the step wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, and the one or more hardware processors are configured to output the adjacency matrix used for the post-update model or a difference matrix between the adjacency matrix used for the pre-update model and the adjacency matrix used for the post-update model. See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) containing "d" nodes. The system may generate an adjacency matrix based on observational input data that indicates the causal relations between the plurality of variables, i.e. wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, and the one or more hardware processors are configured to output the adjacency matrix used for the post-update model (e.g. the adjacency matrix is referred to as an output in [0056]).)
Regarding dependent claim 20,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses an information processing system comprising:the information processing device according to claim 1; See Paragraph [0020], (Data processing apparatus 20 comprises a computing apparatus 22 configured to obtain datasets from a data store 30, i.e. the information processing device according to claim 1;)
Additionally, HORIWAKI further discloses a change detection device configured to detect a change in a state of a target system configured to output the one or more pieces of record data, wherein the change detection device is configured to cause the information processing device to execute processing when detecting a change in the state of the target system. See Paragraph [0015], (The causal relation model verification system may detect mutual contrarieties by performing comparison processing on a causal relation model and domain knowledge data, i.e. a change detection device configured to detect a change in a state of a target system configured to output the one or more pieces of record data, wherein the change detection device is configured to cause the information processing device to execute processing when detecting a change in the state of the target system.)
Regarding dependent claim 21,
As discussed above with claim 20, SANCHEZ-HORIWAKI discloses all of the limitations.
HORIWAKI further discloses the step wherein the target system is a system that manufactures a product, the plurality of variables include a quality characteristic of the product as a variable, See Paragraph [0069], (Control section 30 may process monitor data acquired from facilities and sensors associated with a manufacturing process. Quality data is acquired from facilities and sensors during the quality testing process, i.e. wherein the target system is a system that manufactures a product, the plurality of variables include a quality characteristic of the product as a variable.)
and the change detection device is configured to cause the information processing device to execute processing when a record value of the variable representing the quality characteristic of the product deviates from control limits or specification limits defined by a control chart generated in advance. See FIG. 5 & Paragraph [0051]l, (FIG. 5 illustrates the method comprising step S3006 wherein the system may verify that the subset of monitor data does not contradict domain knowledge, i.e. the change detection device is configured to cause the information processing device to execute processing when a record value of the variable representing the quality characteristic of the product deviates from control limits or specification limits.) See Paragraph [0032], (The system acquires a causal relation model construction condition which indicates a case where a causal relation model is constructed, which includes acquiring domain knowledge of a target manufacturing process, i.e. a control chart generated in advance (e.g. domain knowledge of the target manufacturing process exists prior to the creation of the causal relation model).)
Regarding independent claim 22,
The claim is analogous to the subject matter of independent claim 1 directed to a method or process and is rejected under similar rationale.
Regarding independent claim 23,
The claim is analogous to the subject matter of independent claim 1 directed to a computer system and is rejected under similar rationale.
Claim(s) 6 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ in view of HORIWAKI as applied to claim 1 above, and further in view of KOBAYASHI et al. (WIPO Invention Application Publication No.: WO 2024/117019 A1; Publication Date: June, 6, 2024).
Regarding dependent claim 6,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses the step wherein the structural causal model is represented using an adjacency matrix representing influence from one variable to another variable for each combination of two variables in the plurality of variables, See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) containing "d" nodes. The system may generate an adjacency matrix based on observational input data that indicates the causal relations between the plurality of variables, i.e. the structural causal model is represented using an adjacency matrix representing influence from one variable to another variable for each combination of two variables in the plurality of variables.)
SANCHEZ-HORIWAKI does not disclose the step wherein the one or more hardware processors are configured to: estimate a new value corresponding to a nonzero element in the adjacency matrix of the pre-update model using linear regression based on the one or more pieces of record data when determining that the causal relationships are different, and generate the post-update model by updating a value of a nonzero element in the adjacency matrix of the pre-update model to the estimated new value.
KOBAYASHI discloses the step wherein the one or more hardware processors are configured to: estimate a new value corresponding to a nonzero element in the adjacency matrix of the pre-update model using linear regression based on the one or more pieces of record data when determining that the causal relationships are different, and generate the post-update model by updating a value of a nonzero element in the adjacency matrix of the pre-update model to the estimated new value. See Pg. 6, Paragraph 8, (Disclosing a system for deriving a causal structure of observed variables in an observation system. The method of deriving a causal structure includes representing the relationship between observed variables as a directed acyclic graph having a corresponding adjacency matrix. The adjacency matrix comprises linear regression coefficients that represent the strength of a connection between a pair of observed variables.) See Pg. 3, Paragraph 6, (Control unit 101 optimizes the linear regression coefficient by repeating regression analysis and evaluating the independence of regression residuals.) See Pg. 8, Paragraphs 9-10, (Control unit 101 may recalculate an influence degree based on modifications to the causal structure using linear regression, i.e. estimate a new value corresponding to a nonzero element in the adjacency matrix of the pre-update model using linear regression based on the one or more pieces of record data when determining that the causal relationships are different, and generate the post-update model by updating a value of a nonzero element in the adjacency matrix of the pre-update model to the estimated new value (e.g. the causal structure is updated by the recalculation of influence degrees performed in response to a change in the causal structure such as the addition of new edges or deletion of unnecessary edges.).)
SANCHEZ-HORIWAKI and KOBAYASHI are analogous art because they are in the same field of endeavor, information processing. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of determining causal relationships between variables and displaying the causal relationships to a user as disclosed by KOBAYASHI. Pg. 7, Paragraphs 1-2 of KOBAYASHI disclose that the causal relationships between observed variables may be presented to the user and allows for edge pruning to prevent multiple collinear nodes from simultaneously having a causal relationship with other nosed. This prevents misidentification of edges and enables more reliable learning of causal structures.
Claim(s) 7-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ in view of HORIWAKI as applied to claim 1 above, and further in view of CAO (China Invention Application Publication No.: CN 118018429 A; Pub. Date: May 10, 2024).
Regarding dependent claim 7,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses the step wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) containing "d" nodes. The system may generate an adjacency matrix based on observational input data that indicates the causal relations between the plurality of variables, i.e. the structural causal model is represented using an adjacency matrix representing influence from one variable to another variable for each combination of two variables in the plurality of variables.)
SANCHEZ-HORIWAKI does not disclose the step wherein the one or more hardware processors are configured to generate the post-update model using regularized linear regression based on the one or more pieces of record data while causing a causal order to be the same as the adjacency matrix of the pre-update model, when determining that the causal relationships are different.
Regarding dependent claim 8,
As discussed above with claim 7, SANCHEZ-HORIWAKI-CAO discloses all of the limitations.
CAO further discloses the step wherein the one or more hardware processors are configured to generate the post-update model using Lasso or Adaptive Lasso based on the one or more pieces of record data so as to cause the causal order of the post-update model to be the same as the causal order of the adjacency matrix of the pre-update model, when determining that the causal relationships are different. See Pg. 3, Paragraphs 4-13, (Disclosing a system for estimating a node topological order to optimize network topological efficiency. The method comprises steps S101-S105 wherein step S104 comprises performing Lasso estimation on non-zero elements of a weighted adjacent matrix of a Bayesian network model according to a second data matrix to obtain element weight values of the weighted adjacent metrics. At step S105, the system constructs a directed acyclic graph (DAG) of network service satisfaction metrics according to the weighted adjacency metrics, i.e. wherein the one or more hardware processors are configured to generate the post-update model using Lasso or Adaptive Lasso (e.g. the Lasso estimation is performed on non-zero elements of a weighted adjacent matrix). Lasso estimation is performed on a weighed adjacent matrix associated with a node topological sequence, i.e. based on the one or more pieces of record data while causing a causal order to be the same as the adjacency matrix of the pre-update model, when determining that the causal relationships are different (e.g. the second data matrix comprises a reordering of the data matrix of the pre-trained Bayesian network model).)
SANCHEZ, HORIWAKI and CAO are analogous art because they are in the same field of endeavor, information processing. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of determining causal relationships between variables via the methods described by CAO. Pg. 12, Paragraph 10 of CAO discloses an embodiment wherein performing Lasso estimation on non-zero elements of a dataset according to an optimized matrix represents a process of continuous optimization which improves the DAG learning efficiency.
Regarding dependent claim 9,
As discussed above with claim 7, SANCHEZ-HORIWAKI-CAO discloses all of the limitations.
CAO further discloses the step wherein the one or more hardware processors are configured to generate the post-update model using any one of Adaptive Lasso, Transfer Lasso, or Adaptive Transfer Lasso with the pre-update model as an initial estimation amount, based on the one or more pieces of record data so as to cause the causal order to be the same as the adjacency matrix of the pre-update model, when determining that the causal relationships are different. See Pg. 3, Paragraphs 4-13, (Disclosing a system for estimating a node topological order to optimize network topological efficiency. The method comprises steps S101-S105 wherein step S104 comprises performing Lasso estimation on non-zero elements of a weighted adjacent matrix of a Bayesian network model according to a second data matrix to obtain element weight values of the weighted adjacent metrics. At step S105, the system constructs a directed acyclic graph (DAG) of network service satisfaction metrics according to the weighted adjacency metrics. Note Pg. 8, Paragraph 7 wherein the Lasso estimation is described as corresponding to Adaptive Lasso, i.e. generate the post-update model using any one of Adaptive Lasso, Transfer Lasso, or Adaptive Transfer Lasso with the pre-update model as an initial estimation amount (e.g. the Lasso estimation is performed on non-zero elements of a weighted adjacent matrix). Lasso estimation is performed on a weighed adjacent matrix associated with a node topological sequence, i.e. based on the one or more pieces of record data so as to cause the causal order to be the same as the adjacency matrix of the pre-update model, when determining that the causal relationships are different (e.g. the second data matrix comprises a reordering of the data matrix of the pre-trained Bayesian network model).)
SANCHEZ, HORIWAKI and CAO are analogous art because they are in the same field of endeavor, information processing. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of determining causal relationships between variables via the methods described by CAO. Pg. 12, Paragraph 10 of CAO discloses an embodiment wherein performing Lasso estimation on non-zero elements of a dataset according to an optimized matrix represents a process of continuous optimization which improves the DAG learning efficiency.
Claim(s) 11-12 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ in view of HORIWAKI as applied to claim 10 above, and further in view of UEMURA et al. (WIPO PCT Application No. WO 2024180746 A1; Date Filed: March, 1, 2023).
Regarding dependent claim 11,
As discussed above with claim 10, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ-HORIWAKI does not disclose the step discloses the step wherein the one or more hardware processors are configured to generate the post-update model based on a causal discovery method based on non-Gaussianity.
UEMURA discloses the step wherein the one or more hardware processors are configured to generate the post-update model based on a causal discovery method based on non-Gaussianity. See Pg. 7, Paragraph 8, (Disclosing an information processing device configured to determine the presence or absence of independence or uncorrelatedness between two variables for each of a plurality of combinations. Information processing device 10 applies DirectLiNGAM to a plurality of variables x1, x2, x3, x4, x5 to execute a causal search process.) See Pg. 11, Paragraph 10, (Causal search unit 30 evaluates the independence between a most upstream variable and the remaining other variables as part of the causal search process and subsequently updates each variable in a first group to be excluded and continues identifying the most upstream variable from each updated variable of the first group and each un-updated variable of a second group, , i.e. wherein the one or more hardware processors are configured to generate the post-update model based on a causal discovery method based on non-Gaussianity.)
The examiner notes that one of ordinary skill in the art would recognize that LiNGAM is an abbreviation for "Linear, Non-Gaussian, Acyclic causal Models", DirectLiNGAM being a variation of LiNGAM represents a "causal discovery method based on non-Gaussianity".
SANCHEZ, HORIWAKI and UEMURA are analogous art because they are in the same field of endeavor, information processing. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of determining causal relationships between variables via the methods described by UEMURA. Pg. 12, Paragraph 9 of UEMURA discloses that the information processing system may improve the causal search process by increasing the number of samples relative to the number of variables, thereby improving the accuracy of identification and effectiveness of preventing the algorithm from stopping.
Regarding dependent claim 12,
As discussed above with claim 11, SANCHEZ-HORIWAKI-UEMURA discloses all of the limitations.
UEMURA further discloses the step wherein the one or more hardware processors are configured to generate the post-update model based on ICA-LiNGAM or DirectLiNGAM. See Pg. 7, Paragraph 8, (Disclosing an information processing device configured to determine the presence or absence of independence or uncorrelatedness between two variables for each of a plurality of combinations. Information processing device 10 applies DirectLiNGAM to a plurality of variables x1, x2, x3, x4, x5 to execute a causal search process.) See Pg. 11, Paragraph 10, (Causal search unit 30 evaluates the independence between a most upstream variable and the remaining other variables as part of the causal search process and subsequently updates each variable in a first group to be excluded and continues identifying the most upstream variable from each updated variable of the first group and each un-updated variable of a second group, , i.e. herein the one or more hardware processors are configured to generate the post-update model based on ICA-LiNGAM or DirectLiNGAM.)
SANCHEZ, HORIWAKI and UEMURA are analogous art because they are in the same field of endeavor, information processing. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of determining causal relationships between variables via the methods described by UEMURA. Pg. 12, Paragraph 9 of UEMURA discloses that the information processing system may improve the causal search process by increasing the number of samples relative to the number of variables, thereby improving the accuracy of identification and effectiveness of preventing the algorithm from stopping.
Claim(s) 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ in view of HORIWAKI as applied to claim 10 above, and further in view of SUN et al. (US PGPUB No. 2022/0398260; Pub. Date: Dec. 15, 2022).
Regarding dependent claim 11,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses the step wherein the one or more hardware processors are configured to: generate the plurality of exogenous noise estimation values for each of the one or more pieces of record data, based on the one or more pieces of record data and the post-update model; See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) consisting of a collection of assignments of elements of a d-dimensional random vector "x" wherein each element xi of the vector is determined according to a mathematical relationship between parent nodes of xi in the DAG and a noise term independent of xi referred to as exogenous noise, i.e. generate the plurality of exogenous noise estimation values for each of the one or more pieces of record data, based on the one or more pieces of record data and the post-update model (e.g. noise metrics are determined for a current version of a DAG representing causal relationships);)
SANCHEZ-HORIWAKI does not disclose the step of output[ting] at least one piece of information from information indicating whether there is a change in causal influence, whether there is a change in causal structure, and whether there is a change in causal order, based on independence between any two or more variables in the plurality of exogenous noise estimation values after update for each of the one or more pieces of record data.
SUN discloses the step of output[ting] at least one piece of information from information indicating whether there is a change in causal influence, whether there is a change in causal structure, and whether there is a change in causal order, based on independence between any two or more variables in the plurality of exogenous noise estimation values after update for each of the one or more pieces of record data. See FIG. 1 & Paragraph [0116], (Disclosing an information processing method for determining independence relationships from among a plurality of variables. FIG. 1 illustrates the method comprising step S13 of adjusting a first independent relationship according to an adjustment scheme to obtain a second independence relationship. The system may adjust a first independent value so that the correct causal relationship information may be obtained as in step S14. Note [0074] wherein the system may reduce the impact of data noise using a composite score to determine the possibility that an independence relationship is correct, i.e. output at least one piece of information from information indicating whether there is a change in causal influence (e.g. the obtained causal relationship information is an output of the method of FIG. 1), based on independence between any two or more variables in the plurality of exogenous noise estimation values (e.g. the system may determine independence of variables under a condition while reducing the impact of noise via a composite score) after update for each of the one or more pieces of record data (e.g. the process of obtaining a second independent relationship comprises adjusting a first independence relationship according to a first adjustment scheme, i.e. an update.).)
SANCHEZ, HORIWAKI and SUN are analogous art because they are in the same field of endeavor, information processing. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of determining independence between a plurality of variables as disclosed by SUN. Paragraph [0059] of SUN discloses that the method of determining independent relationships and its associated steps of adjusting according to an adjustment scheme allow the system to continuously improve the accuracy of the independence relationship, ensuring the accuracy stability and interpretability of the independent relationship output even in the case of low data quality.
Claim(s) 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ in view of HORIWAKI as applied to claim 10 above, and further in view of MUELLER et al. (US PGPUB No. 2021/0256406; Pub. Date: Aug. 19, 2021).
Regarding dependent claim 18,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses the step wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables, See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) containing "d" nodes. The system may generate an adjacency matrix based on observational input data that indicates the causal relations between the plurality of variables, i.e. wherein the structural causal model is represented using an adjacency matrix representing a magnitude of influence from one variable to another variable for each combination of two variables in the plurality of variables)
SANCHEZ-HORIWAKI does not disclose the step wherein the one or more hardware processors are configured to display a causal model representing the adjacency matrix used for the structural causal model after update.
MUELLER discloses the step wherein the one or more hardware processors are configured to display a causal model representing the adjacency matrix used for the structural causal model after update. See FIG. 1 & Paragraph [0100], (Disclosing a system for generating an interactive visualization of causal models used in data analytics. FIG. 1 illustrates a causal network visualization interface. Users may start form either a causality model or correlation graph as shown in FIG. 1A. Note [0115] wherein each causal graph may be represented as an adjacency matrix, i.e. the one or more hardware processors are configured to display a causal model representing the adjacency matrix used for the structural causal model after update (e.g. Note [0119] wherein adjacency matrices may be clustered to uncover different causal mechanisms, which allows the system to update the causal model and data subdivisions to draw updated causal models, model heatmaps and/or model similarity plots).)
SANCHEZ, HORIWAKI and MUELLER are analogous art because they are in the same field of endeavor, causal modelling. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of visualizing causal models as disclosed by MUELLER. Paragraph [0076] of MUELLER discloses that the visual analytics system provides improved processing and handling of heterogeneous data in causal inference with its experimental evaluation. The plurality of interactive function allow users to explore sub-divisions of data from which different models may be inferred.
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over SANCHEZ in view of HORIWAKI as applied to claim 10 above, and further in view of MUELLER et al. (US PGPUB No. 2021/0256406; Pub. Date: Aug. 19, 2021).
Regarding dependent claim 18,
As discussed above with claim 1, SANCHEZ-HORIWAKI discloses all of the limitations.
SANCHEZ further discloses an information processing system comprising: the information processing device according to claim 1; See Paragraph [0020], (Data processing apparatus 20 comprises a computing apparatus 22 configured to obtain datasets from a data store 30, i.e. the information processing device according to claim 1;)
and an analysis device, wherein the analysis device is configured to: calculate the plurality of exogenous noise estimation values based on the one or more pieces of record data and the structural causal model; See Paragraph [0056], (The process of discovering the causal structure between variables includes defining a structural causal model (SCM) embodied as a directed acyclic graph (DAG) consisting of a collection of assignments of elements of a d-dimensional random vector "x" wherein each element xi of the vector is determined according to a mathematical relationship between parent nodes of xi in the DAG and a noise term independent of xi referred to as exogenous noise.
SANCHEZ-HORIWAKI does not disclose the step of generat[ing] a degree of contribution representing a magnitude of influence that the exogenous noise given to a source variable that is a first one of two variables exerts on a target variable that is a second one of the two variables for each combination of the two variables in the plurality of variables in each of the one or more pieces of record data, based on the structural causal model and the plurality of exogenous noise estimation values for each of the one or more pieces of record data.
Perov discloses the step of generat[ing] a degree of contribution representing a magnitude of influence that the exogenous noise given to a source variable that is a first one of two variables exerts on a target variable that is a second one of the two variables for each combination of the two variables in the plurality of variables in each of the one or more pieces of record data, based on the structural causal model and the plurality of exogenous noise estimation values for each of the one or more pieces of record data. See Paragraph [0028], (Disclosing a system for performing inference on a generative model in a probabilitic program form configured to define variables and probabilistic relationships between said variables. Noise is modelled as a separate variable in the structured causal model. Note [0121] wherein the structured causal model separate variables into exogenous and endogenous variables.) See FIG. 4C & Paragraph [0122], (FIG. 4C illustrates a scenario where a noise value is shown for two queries wherein the explicit noise variables are labelled Y_N_1 and Y_N_2, i.e. generate a degree of contribution representing a magnitude of influence that the exogenous noise given to a source variable that is a first one of two variables exerts on a target variable that is a second one of the two variables for each combination of the two variables in the plurality of variables in each of the one or more pieces of record data, based on the structural causal model and the plurality of exogenous noise estimation values for each of the one or more pieces of record data.
SANCHEZ, HORIWAKI and Perov are analogous art because they are in the same field of endeavor, causal modelling. It would have been obvious to anyone having ordinary skill in the art before the effective filing date to modify the system of SANCHEZ-HORIWAKI to include the method of performing structured causal modelling as disclosed by Perov. Paragraph [0106] of Perov discloses that the approximate interference method of importance sampling that is beneficial in inference where it may be difficult or computationally expensive to generate samples from the real posterior. Sampling from the proposal can make the inference more efficient and improve the resource-burden of computing inference queries.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Fernando M Mari whose telephone number is (571)272-2498. The examiner can normally be reached Monday-Friday 7am-4pm.
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, Ann J. Lo can be reached at (571) 272-9767. 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.
/FMMV/Examiner, Art Unit 2159
/ANN J LO/Supervisory Patent Examiner, Art Unit 2159