DETAILED ACTION
This action is made FINAL in response to the amendments filed on 6/24/2026.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 – 30 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step One
The claims are directed to a method (claims 1 - 9), an apparatus with structural components (claims 10 - 27), and a non-transitory computer readable medium (claims 28 - 30). Thus, each of the claims falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
As to claims 1,
Step 2A, Prong One
The claim recites in part:
determining a probability distribution for each variable of the plurality of
variables based on the observation data;
For example, a human can observe past outcomes and estimates the probability distribution of each variable by tracking how frequently each outcome occurs over time.
computing a likelihood of including each edge in a causal graph based on the probability distribution and the intervention data, each edge representing a direct causal relation between variables of the plurality of variables, the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter modeling
existence of the edge and the second parameter modeling a direction of the edge;
For example, a human reviews observed data and interventions and estimates the likelihood of causal link (edge) by comparing how often variables change together versus independent
generating the causal graph based on the likelihood of including each edge.
For example, a person uses a pencil and paper to count how often variables occur together, then draws connections between them based on frequencies. Humans have been created graphs before computers where ever invented.
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The judicial exception is not integrated into a practical application. In particular, the claim recites the additional elements of:
receiving a data set including observation data and intervention data corresponding to a plurality of variables;
which amounts to extra-solution activity of gathering data for use in the claimed process. As described in MPEP 2106.05(g), limitations that amount to merely adding insignificant extra-solution activity to a judicial exception do not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application.
The recitation of observation data, intervention data, and edge amounts to generally linking the use of the judicial exception to a particular environment of field of use (See MPEP 2106.05(h)).
Accordingly, at Step 2A, Prong Two, the additional elements individually or in combination do no integrate the judicial exception into a practical application.
Step 2B
In accordance with Step 2B, the claim does not include additional elements that are sufficient to amount to significantly more that the judicial exception. As discussed above, the additional elements of:
receiving a data set including observation data and intervention data corresponding to a plurality of variables;
are recited at a high level of generality and amounts to extra-solution activity of receiving data i.e. pre-solution activity of gathering data for use in the claimed process. The courts have found limitations directed to obtaining information electronically, recited at a high level of generality, to be well-understood, routine, and conventional (see MPEP 2106.05(d)(II), “receiving or transmitting data over a network”, "electronic record keeping," and "storing and retrieving information in memory").
The recitation of observation data, intervention data, and edge amounts to generally linking the use of the judicial exception to a particular environment of field of use (See MPEP 2106.05(h)).
Accordingly, at Step 2B the additional elements individually or in combination do not amount to significantly more than the judicial exception.
As to claim 2,
Step 2A, Prong One
The claim recites in part:
updating the graph by iteratively repeating the determining the probability distribution and computing the likelihood of including each edge.
For example, a human can repeatedly update the graph using a pencil and paper. Humans have been created graphs and updated graphs before computers where ever invented.
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 3,
Step 2A, Prong One
The claim recites in part:
in which the probability distribution is determined by dropping one or more variables as inputs.
For example, a human can mentally remove a variable and not include it when determining the probability distribution.
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 4,
Step 2A, Prong One
The claim recites in part:
in which the dropping is performed mentally
For example, a human can randomly select and mentally remove a variable and not include it when determining the probability distribution.
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 5,
Step 2A, Prong One
The claim recites in part:
in which the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter models existence of an edge and the second parameter models a direction of the edge.
For example, a human uses pencil and paper to note which variables occur together and which comes first then draws edges and directions accordingly.
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 6,
Step 2A, Prong One
The claim recites in part:
comparing the likelihood of including each edge to a predefined threshold; and removing an edge from the graph based on the comparison.
For example, a human uses a pencil and paper to compare how often variables are related, crosses out connections that fall below a predetermined threshold, and keeps only the strongest edges in the graph
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 7,
Step 2A, Prong One
The claim recites in part:
wherein the second parameter controls the graph to be acyclic
For example, a human uses a pencil and paper to draw connections, then avoids or erases any edge that would create a loop, ensuring the graph stays acyclic
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 8,
Step 2A, Prong One
The claim recites in part:
in which the likelihood of including each edge is computed based on a sparsity regularizer
For example, a human uses pencil and paper to prefer a simpler graph by only keeping the strongest relationships and leaving out the weaker ones (sparsity).
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
As to claim 9,
Step 2A, Prong One
The claim recites in part:
outputting a causal graph when the likelihood of including each edge in the graph converges to one
For example, a human uses a pencil and paper to track relationships and once a connection consistently appears every time, the human keeps that edge in the final causal graph.
As drafted and under its broadest reasonable interpretation, these limitation covers performance of the limitation in the mind (including an observation, evaluation, judgment, opinion) or with the aid of pencil and paper but for the recitation of generic computer components.
Accordingly, at Step 2A, Prong One, the claim is directed to an abstract idea.
Step 2A, Prong Two
The claim does not include additional elements that integrate the judicial exception into a practical application or amount to significantly more than the judicial exception itself
Step 2B
The claim does not include additional elements that are sufficient to amount to “significantly more” to the judicial exception
Claim 10 has similar limitations as claim 1. Therefore, the claim is rejected for the same reasons as above.
The claim further recites a memory and at least one processor which are recited at a high-level of generality and amounts to no more than mere instructions to apply the exception using a generic computer component (See MPEP 2106.05(f)).
Claim 11 has similar limitations as claim 2. Therefore, the claim is rejected for the same reasons as above.
Claim 12 has similar limitations as claim 3. Therefore, the claim is rejected for the same reasons as above.
Claim 13 has similar limitations as claim 4. Therefore, the claim is rejected for the same reasons as above.
Claim 14 has similar limitations as claim 5. Therefore, the claim is rejected for the same reasons as above.
Claim 15 has similar limitations as claim 6. Therefore, the claim is rejected for the same reasons as above.
Claim 16 has similar limitations as claim 7. Therefore, the claim is rejected for the same reasons as above.
Claim 17 has similar limitations as claim 8. Therefore, the claim is rejected for the same reasons as above.
Claim 18 has similar limitations as claim 9. Therefore, the claim is rejected for the same reasons as above.
Claim 19 has similar limitations as claim 1. Therefore, the claim is rejected for the same reasons as above.
Claim 20 has similar limitations as claim 2. Therefore, the claim is rejected for the same reasons as above.
Claim 21 has similar limitations as claim 3. Therefore, the claim is rejected for the same reasons as above.
Claim 22 has similar limitations as claim 4. Therefore, the claim is rejected for the same reasons as above.
Claim 23 has similar limitations as claim 5. Therefore, the claim is rejected for the same reasons as above.
Claim 24 has similar limitations as claim 6. Therefore, the claim is rejected for the same reasons as above.
Claim 25 has similar limitations as claim 7. Therefore, the claim is rejected for the same reasons as above.
Claim 26 has similar limitations as claim 8. Therefore, the claim is rejected for the same reasons as above.
Claim 27 has similar limitations as claim 9. Therefore, the claim is rejected for the same reasons as above.
Claim 28 has similar limitations as claim 1. Therefore, the claim is rejected for the same reasons as above.
Claim 29 has similar limitations as claim 2. Therefore, the claim is rejected for the same reasons as above.
Claim 30 has similar limitations as claim 3. Therefore, the claim is rejected for the same reasons as above.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 2, 5, 7, 10, 11, 14, 16 19, 20, 23, 25, 28, and 29 is/are rejected under 35 U.S.C. 103 as being unpatentable over Glass et al (US 2017/0061324) in view of ZHANG et al (US 2023/0229906).
As to claim 1, Glass et al teaches in figures 2 - 4 shows and teaches a method, comprising:
receiving a data set including observation data and intervention data
corresponding to a plurality of variables (paragraph [0005]… The creating includes receiving a training set that includes multiple scenarios, each scenario comprised of one or more natural language statements, and each scenario corresponding to a plurality of candidate answers ; paragraph [0044]…training set 402 is received) (Examiner’s Note: “receiving a training set that includes multiple scenarios, each scenario comprised of one or more natural language statements, and each scenario corresponding to a plurality of candidate answers” reads on “receiving a data set including observation data and intervention data corresponding to a plurality of variables” ; “one or more natural language statements” reads on “observation data” ; “plurality of candidate answers” reads on “intervention data” ; “each scenario” reads on “plurality of variables”)
determining a probability distribution for each variable of the plurality of
variables based on the observation data (paragraph [0022]….The term “inference graph” as used herein refers to any graph represented by a set of nodes connected by edges, where the nodes represent statements and the edges represent relations between . An inference graph can be used to represent relation paths between factors in an inquiry and possible answer to that inquiry. An inference graph is multi-step if it contains more than one edge in a path from a set of factors to an answer. In an embodiment, graph nodes, edges/attributes (confidences), statements and relations may be represented in software, as Java objects. Confidences, strengths, and probabilities are attached to them for processing by various computer systems ; paragraph [0049]… a process is performed where features based on the raw confidence from the inference model are transformed into a proper probability distribution over the candidate answers) (Examiner’s Note: “a process is performed where features based on the raw confidence from the inference model are transformed into a proper probability distribution” reads on “determining a probability distribution for each variable of the plurality of variables based on the observation data”) ;
computing a likelihood of including each edge in a causal graph based on the probability distribution and the intervention data, each edge connecting variables of the plurality of variables (paragraph [0023]… embodiments include answering scenario questions by constructing a graph that links factors extracted from the scenario (e.g., symptoms or lab values) to candidate diagnoses. Edges can be added to this graph by using a factoid QA subsystem to answer questions about existing nodes. For example, the system may ask what diseases cause particular symptoms. Each edge has a confidence value and a vector of features assigned by the underlying factoid QA system. As used herein, the term “feature” refers to a number summarizing a relevant aspect of the task. As used herein, the term “vector of features” refers to any collection of features. Embodiments estimate the confidence in each candidate diagnosis, which depends on the strength and structure of its connection to the factors)(Examiner’s Note: ”Each edge has a confidence value” reads on “computing a likelihood of including each edge in a graph” ; “Edges can be added to this graph by using a factoid QA subsystem to answer questions about existing nodes” reads on “each edge connecting variables of the plurality of variables”); and
generating the graph based on the likelihood of including each edge (paragraph [0005]… The creating includes receiving a training set that includes multiple scenarios, each scenario comprised of one or more natural language statements, and each scenario corresponding to a plurality of candidate answers. The creating also includes constructing evidence graphs for each of the multiple scenarios based on the training set, and calculating weights for common features across the evidence graphs that will maximize a probability of the inference model locating correct answers from corresponding candidate answers across all of the multiple scenarios. In response to an inquiry from a user via the computer processor, where the inquiry includes a scenario, the inference model constructs an evidence graph and recursively constructs formulas to express a confidence of each node in the evidence graph in terms of its parents in the evidence graph ; paragraph [0065]… Each inference model is also trained on the full training set, these versions are applied at test time to generate the features for the ensemble. Referring back to FIG. 4, at block 408, an inference model is generated from the results of steps 402-406. The inference model is ready for use by a question answering system) (Examiner’s Note: “constructing evidence graphs” reads on “generating a graph”).
Glass et al fails to explicitly show/teach each edge representing a direct causal relation between variables of the plurality of variables, the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter modeling existence of the edge and the second parameter modeling a direction of the edge.
However, ZHANG et al teaches each edge representing a direct causal relation between variables of the plurality of variables, the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter modeling existence of the edge and the second parameter modeling a direction of the edge (paragraph [0009]… According to one aspect disclosed herein, there is provided computer-implemented method comprising accessing a machine learning, ML, model that is operable to sample a causal graph from a graph distribution describing different possible graphs. In a causal graph, nodes represent the different variables of said set and edges represent causation, and the graph distribution comprises a matrix of probabilities of existence and causal direction of potential edges between pairs of nodes. The ML model has been pre-trained to be able to generate a respective simulated value of a selected variable from among said set based on the sampled causal graph. The method further comprises using the (trained) ML model to estimate a treatment effect from one or more intervened-on variables on another, target variable from among the variables of said set. This is done by: a) selecting the target variable as the selected variable to be simulated by the ML model; b) fixing the input value of each intervened-on variable to a specified value, including disregarding any edge directed from any parent of the intervened-on variable into the intervened-on variable in the sampled causal graph; c) sampling a causal graph from the graph distribution and observing the corresponding simulated value of the target variable; d) repeating c) multiple times, re-sampling the causal graph from the graph distribution each time; and e) determining an expectation of the target variable by averaging the simulated values of the target variable from c)-d) over the multiple sampled graphs, thus giving the estimated treatment effect) (Examiner’s Note: “In a causal graph, nodes represent the different variables of said set and edges represent causation, and the graph distribution comprises a matrix of probabilities of existence and causal direction of potential edges between pairs of nodes.” reads on “each edge representing a direct causal relation between variables of the plurality of variables, the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter modeling existence of the edge and the second parameter modeling a direction of the edge”)
Therefore, it would have been obvious for one having ordinary skill in the art, at the time the invention was made, for Glass et al’s each edge representing a direct causal relation between variables of the plurality of variables, the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter modeling existence of the edge and the second parameter modeling a direction of the edge, as in ZHANG et al, for the purpose of being effective for performing machine learning tasks.
As to claim 2, Glass et al teaches in figures 2 - 4 shows the method, further comprising updating the graph by iteratively repeating the determining the probability distribution and computing the likelihood of including each edge (paragraph [0069] Confidences in the nodes are estimated 512 and a hypothesis is identified 514 from the confidences. The exemplary embodiments, as described above, learn parameter values in support of the estimation of the confidences in the nodes 512. The hypothesis 514 may then be fed back to the assertion graph 506, which then re-prioritizes the nodes 508 based on updates to the assertion graph 506. The process cycle from block 514, block 506, and block 508 may be repeated until, e.g., a desired outcome or solution is reached)(Examiner’s Note: “The process cycle from block 514, block 506, and block 508 may be repeated until, e.g., a desired outcome or solution is reached” reads on “updating the graph by iteratively repeating the determining the probability distribution and computing the likelihood of including each edge”).
As to claim 5, Glass et al teaches in figures 2 - 4 shows the method, in which the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter models existence of an edge and the second parameter models a direction of the edge (paragraph [0033]… a directed acyclic graph (DAG) 200 is generated from an evidence graph, whereby the evidence graph is transformed into a feed-forward model in which matching nodes are clustered and cycles are broken by re-orienting edges to point from factors to candidates, producing a directed acyclic graph (DAG). A layer 202, shown between the factors 104 and candidate answers 106, illustrates this aspect. The confidence in clinical factors (factors 104) is 1.0. For all other nodes the confidence is defined recursively in terms of the confidences of the parents and the confidence of the edges produced by the system. Let the set of parents for a node n be given by a(n). The confidence the QA system gives for one node, m, indicating another, n, is given by c(m,n). Then the confidence for non-clinical factor nodes is given below)(Examiner’s Note: “which matching nodes are clustered” reads on “the first parameter the first parameter models existence of an edge” ; “cycles are broken by re-orienting edges to point from factors to candidates” reads on “the second parameter models a direction of the edge”).
As to claim 7, Laszlo et al teaches the method, in which the second parameter controls the graph to be acyclic (paragraph [0033]… a directed acyclic graph (DAG) 200 is generated from an evidence graph, whereby the evidence graph is transformed into a feed-forward model in which matching nodes are clustered and cycles are broken by re-orienting edges to point from factors to candidates, producing a directed acyclic graph (DAG))
Claim 10 has similar limitations as claim 1. Therefore, the claim is rejected for the same reasons as above.
Claim 11 has similar limitations as claim 2. Therefore, the claim is rejected for the same reasons as above.
Claim 14 has similar limitations as claim 5. Therefore, the claim is rejected for the same reasons as above.
Claim 16 has similar limitations as claim 7. Therefore, the claim is rejected for the same reasons as above.
Claim 19 has similar limitations as claim 1. Therefore, the claim is rejected for the same reasons as above.
Claim 20 has similar limitations as claim 2. Therefore, the claim is rejected for the same reasons as above.
Claim 23 has similar limitations as claim 5. Therefore, the claim is rejected for the same reasons as above.
Claim 25 has similar limitations as claim 7. Therefore, the claim is rejected for the same reasons as above.
Claim 28 has similar limitations as claim 1. Therefore, the claim is rejected for the same reasons as above.
Claim 29 has similar limitations as claim 2. Therefore, the claim is rejected for the same reasons as above.
Claim(s) 3, 4, 12, 13, 15, 21, 22, 24, and 30 is/are rejected under 35 U.S.C. 103 as being unpatentable over Glass et al (US 2017/0061324) in view of ZHANG et al (US 2023/0229906) and in further view of Laszlo et al (US 2021/0201158).
As to claim 3, Glass et al teaches the determination of the probability distribution
Glass et al and ZHANG et al both fail to explicitly show/teach that probability distribution is determined by dropping one or more variables as inputs.
However, Laszlo et al teaches that a probability distribution is determined by dropping one or more variables as inputs (paragraph [0237]…the evolutionary system 812 may mutate a graph by applying one or more random modifications to the graph. The random modifications may include, e.g., adding or removing edges between randomly selected pairs of nodes in the graph, or adding random “noise” values (e.g., sampled from a predefined probability distribution) to the weight values associated with the edges of the graph) (Examiner’s Note: “random modifications may include, e.g., adding or removing edges between randomly selected pairs of nodes in the graph” reads on “that probability distribution is determined by dropping one or more variables as inputs”).
Therefore, it would have been obvious for one having ordinary skill in the art, at the time the invention was made, for Glass et al’s probability distribution to be determined by dropping one or more variables as inputs, as in Laszlo et al, for the purpose of being effective for performing machine learning tasks.
As to claim 4, Laszlo et al teaches the method in which the dropping is performed randomly (paragraph [0237]…the evolutionary system 812 may mutate a graph by applying one or more random modifications to the graph. The random modifications may include, e.g., adding or removing edges between randomly selected pairs of nodes in the graph, or adding random “noise” values (e.g., sampled from a predefined probability distribution) to the weight values associated with the edges of the graph).
It is obvious for dropping to be performed randomly, for the same reasons as above.
As to claim 6, Laszlo et al teaches comparing the likelihood of including each edge to a predefined threshold; and removing an edge from the graph based on the comparison.(paragraph [0231]… the graph update engine 820 may generate a candidate graph 802 by iteratively updating an “initial” graph 826, e.g., by adding or removing nodes or edges from the initial graph 826 at each of one or more iterations. The initial graph 826 may be, e.g., a default (predefined) graph, or a randomly generated graph. At each iteration, the graph update engine 820 may update the current graph to cause the current graph to satisfy a corresponding constraint 824 (Examiner’s Note: “the graph update engine 820 may generate a candidate graph 802 by iteratively updating an “initial” graph 826, e.g., by adding or removing nodes or edges from the initial graph 826 at each of one or more iterations” reads on “comparing the likelihood of including each edge to a predefined threshold an removing an edge from the graph based on the comparison” ; corresponding constraint” reads on “predefined threshold”)
It would have been obvious for comparing the likelihood of including each edge to a predefined threshold; and removing an edge from the graph based on the comparison, for the same reasons as above.
Claim 12 has similar limitations as claim 3. Therefore, the claim is rejected for the same reasons as above.
Claim 13 has similar limitations as claim 4. Therefore, the claim is rejected for the same reasons as above.
Claim 15 has similar limitations as claim 6. Therefore, the claim is rejected for the same reasons as above.
Claim 21 has similar limitations as claim 3. Therefore, the claim is rejected for the same reasons as above.
Claim 22 has similar limitations as claim 4. Therefore, the claim is rejected for the same reasons as above.
Claim 30 has similar limitations as claim 3. Therefore, the claim is rejected for the same reasons as above.
Claim 24 has similar limitations as claim 6. Therefore, the claim is rejected for the same reasons as above.
Claim(s) 8, 17, and 26 is/are rejected under 35 U.S.C. 103 as being unpatentable over Glass et al (US 2017/0061324) in view of ZHANG et al (US 2023/0229906) and in further view of Lee et al (US 2019/0065818).
As to claim 8, Glass et all teaches the likelihood of included each edge.
Glass and ZHANG et al both fail to explicitly show/teach that the likelihood of including each edge is computer based on a sparsity regularizer.
However, Lee et al teaches that the likelihood of including each edge is computer based on a sparsity regularizer. (paragraph [0087]…Edge locations may be modeled with a sparsity-inducing probability distribution with a parameter controlling the spread (i.e., variance) of the distribution. When previous data gave evidence for an edge in a given location, the parameter to increase the variance of the prior data in this location, thereby likely making it easier for the inference to identify the presence of the edge from limited observations. In contrast, when previous data indicated that an edge in a location was unlikely, this variance may be decreased thereby requiring more evidence from the observations to infer the presence of an edge) (Examiner’s Note: “Edge locations may be modeled with a sparsity-inducing probability distribution with a parameter controlling the spread (i.e., variance) of the distribution” reads on “the likelihood of including each edge is computer based on a sparsity regularizer”)
Therefore, it would have been obvious for one having ordinary skill in the art, at the time the invention was made, for Glass et al’s likelihood of including each edge is computer based on a sparsity regularizer, as in Lee et al, for the purpose of making it easier for the inference to identify the presence of the edge from limited observations
Claim 17 has similar limitations as claim 8. Therefore, the claim is rejected for the same reasons as above.
Claim 26 has similar limitations as claim 8. Therefore, the claim is rejected for the same reasons as above.
Claim(s) 9, 18, and 27 is/are rejected under 35 U.S.C. 103 as being unpatentable over Glass et al (US 2017/0061324) in view of ZHANG et al (US 2023/0229906) and in further view of Nachmanson et al (US 2005/0160404).
As to claim 9, Glass et all teaches the likelihood of included each edge.
Glass and ZHANG et al both fails to explicitly show/teach outputting a casual graph when the likelihood of including each edge in the graph converges to one.
However, Nachmanson et al teaches outputting a casual graph when the likelihood of including each edge in the graph converges to one.(paragraph [0025]… In one example, a method traditionally used with deterministic graph traversal is considered with probabilistic adaptations. First, in order to cover as many edges of the graph as possible, a Chinese Postman tour algorithm provides a minimal cost tour covering all edges. Next, the tour of the graph is split into sequences such that no sequence contains choice points (i.e., non-deterministic vertices or states) as an internal point. If a player covers all sequences playing the game, then all edges are covered. A string starting from each non-deterministic choice is provided. The environment chooses non-deterministically an edge exiting the selected vertex. The player then follows any sequence starting at the edge destination and continues play from the non-deterministically selected vertex. In one example, the probability to cover all edges converges to one geometrically when the number of steps in a strategy approaches infinity)(Examiner’s Note: “The environment chooses non-deterministically an edge exiting the selected vertex…the probability to cover all edges converges to one geometrically when the number of steps in a strategy approaches infinity” reads on “outputting a casual graph when the likelihood of including each edge in the graph converges to one”).
Therefore, it would have been obvious for one having ordinary skill in the art, at the time the invention was made, for Glass et al’s outputting a casual graph when the likelihood of including each edge in the graph converges to one, as in Nachmanson et al, for the purpose of creating test coverage for non-deterministic programs.
Claim 18 has similar limitations as claim 9. Therefore, the claim is rejected for the same reasons as above.
Claim 27 has similar limitations as claim 9. Therefore, the claim is rejected for the same reasons as above.
Response to Arguments
Applicant's arguments filed 6/24/2026 have been fully considered but they are not persuasive.
Specification
The updated title of the invention overcomes the specification objection and the specification objection is withdrawn.
Claim Rejections - 35 USC § 101
The 101 Rejection still has not been overcome. The claims are abstract and the steps in the claims can be completed with a mental process and/or generic computer components. Additionally, the steps in the claims do not describe an improvement of technology in any way.
The applicant argues:
The Examiner characterizes the claimed operations as performable "in the mind" or "with the aid of pencil and paper." Applicant respectfully submits that this characterization is incorrect. The specification discloses that the claimed process involves Monte Carlo sampling of between 200 and 1,000 graphs per iteration (see specification T[0075]), computing gradients for parameters Yij and Θᵢⱼ according to the mathematical expressions set forth in Equations 2 and 3 of the specification (see specification 11[0073]-[0074]), iterative optimization over all pairs of variables Xᵢ and Xⱼ in the causal graph, and convergence analysis under the limit of infinite data (see specification [0083]). These operations are not practically performable by a human with pencil and paper. The claimed method is directed to a specific computational process for neural causal discovery, which is a technical field requiring significant computational resources. Accordingly, the claims do not recite a mental process and no judicial exception is recited.
The Examiner disagrees. The examiner respectfully disagrees with the applicant’s position, as the arguments presented rely on limitations that are neither explicitly recited in the claims nor reasonably inferred from them. At no point in the pending claims does the applicant assert, describe, or even suggest the limitations of:
that the claimed process involves Monte Carlo sampling of between 200 and 1,000 graphs per iteration
computing gradients for parameters Yij and Θij according to the mathematical expressions set forth in Equations 2 and 3 of the specification
iterative optimization over all pairs of variables Xᵢ and Xⱼ in the causal graph
convergence analysis under the limit of infinite data
Rather, the applicant appears to have introduced this language as part of the argument, but such a limitation cannot be read into the claims when it is not supported by the actual claim language. Without clear support in the claim language the examiner cannot give weight to arguments premised on these alleged limitations. Additionally, the mental process depends on the nature of the claimed analysis, not the quantity of data analyzed. Processing a large number of data items does not transform and otherwise mental process into a non-abstract concept.
The applicant argues:
Even assuming arguendo that a judicial exception is recited, the claims integrate any such exception into a practical application. The specification explicitly identifies and addresses technical problems with conventional causal discovery methods including sensitivity to the weight assigned to the direct acyclic graph (DAG) regularizer, lack of theoretical convergence guarantees, and scalability limitations preventing application to graphs with more than twenty nodes. See, published specification 11 26-30. Amended claims 1, 10, 19, and 28 now expressly recite computing "a likelihood of including each edge in a causal graph based on the probability distribution and the intervention data, each edge representing a direct causal relation between variables of the plurality of variables, where the likelihood of including each edge is determined based on a first parameter and a second parameter, the first parameter modeling existence of the edge and the second parameter modeling a direction of the edge." Thus, unlike conventional approaches, the amended claims enable elimination of a DAG regularizer, provide improved convergence to the true causal graph, and enable scalability to graphs with one thousand or more nodes. See, e.g., published specification, 11 30, 65, and 70. These are concrete technical improvements to the functioning of causal discovery systems, not merely the application of an abstract idea to a generic computer.
Moreover, the claims recite an inventive concept. The Examiner has provided no evidentiary support that the claimed combination of operations including the two-parameter (γ,Θ) edge probability model, the sparsity regularizer, the gradient estimator, and the iterative convergence mechanism is well-understood, routine, or conventional in the field of causal discovery. Under In re Zurko, 258 F.3d 1379 (Fed. Cir. 2001), the USPTO must point to concrete evidence in the record to support such findings. No such evidence has been provided.
The Examiner disagrees. The examiner respectfully disagrees with the applicant’s position, as the arguments presented rely on limitations that are neither explicitly recited in the claims nor reasonably inferred from them. At no point in the pending claims does the applicant assert, describe, or even suggest the limitations of:
sensitivity to the weight assigned to the direct acyclic graph (DAG) regularizer, lack of theoretical convergence guarantees, and scalability limitations preventing application to graphs with more than twenty nodes.
enable elimination of a DAG regularizer, provide improved convergence to the true causal graph, and enable scalability to graphs with one thousand or more nodes.
Rather, the applicant appears to have introduced this language as part of the argument, but such a limitation cannot be read into the claims when it is not supported by the actual claim language. Without clear support in the claim language the examiner cannot give weight to arguments premised on these alleged limitations.
The claims do not got into any detail as to how the current set of claims are even related to a DAG regularizer or provide improved convergence to the true causal graph as this elements are not even mentioned in the claims Additionally, the mental process depends on the nature of the claimed analysis, not the quantity of data analyzed.
Claim Rejections - 35 USC § 102
The newly added limitations overcome the 102 Rejection but do not put the claims in a condition for allowance. As explained above Glass et al in view of ZHANG et al clearly shows/teaches all the limitations as claimed.
Conclusion
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.86t
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRANDON S COLE whose telephone number is (571)270-5075. The examiner can normally be reached Mon - Fri 7:30pm - 5pm EST (Alternate Friday's Off).
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, Omar Fernandez can be reached at 571-272-2589. 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.
/BRANDON S COLE/ Primary Examiner, Art Unit 2128
Prosecution Insights