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 .
Responsive to communication on 04/13/2023
Claims 1-20 pending
Claims 1-20 rejected
Priority
Regarding Application Data Sheet received on 04/13/2023. Application Data sheet does not claim foreign or domestic priority. Application data sheet accepted by the examiner.
Drawings
Responsive to drawings received on 04/13/2023. Drawings are accepted by the examiner.
Specification
Responsive to abstract received on 04/13/2023. Abstract is accepted by the examiner.
Responsive to specifications received on 04/13/2023. Specifications accepted by the examiner.
Claim Interpretation
Claim 2 states “one from among a toward-step computation and an away-step computation.” Claims 3 and 4 state “wherein the away-step computation.” Because the away step computation of claim 2 is not a required limitation, claims 3 and 4 which depend on claim 2 are not required if claim 2 is viewed in respect of a toward-step computation. This interpretation also applies to similar claims 12 and 13.
Claim 6 states “using the identified sparsity pattern to increase a convergence rate,” As understood by the examiner, this refers to a time it takes to converge the MLE, this is not a specific parameter of a function being changed.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention recites a judicial exception, an abstract idea, which has not been integrated into practical application and the claims further do not recite significantly more than the judicial exception.
Claim 1
Step 1: Is the claimed invention one of the four statutory categories? :
YES. The claim recite A method which is a process.
Step 2A Prong 1, inquiry "Is the claim directed to a law of nature, a natural phenomenon or an abstract idea?":
YES. Claim 1 recites: modeling, by the at least one processor based on the received first information, the sequence of events by a multidimensional Hawkes process that relates to a conditional density function that includes a base intensity component and a cross- activation matrix component;
Modeling in this context pertains to the formation of a mathematical model which represents time sequential data. A multidimensional Hawkes process is a mathematic model. A conditional density function is a mathematic function. A base intensity component and cross activation matrix components are mathematic components of the conditional density math function. In this context, modeling a sequence of events is a mathematic process in which a sequence of events is related mathematically for probability predictions to be made. The MPEP 2106.04(a)(2)(I)(B) states “A claim that recites a numerical formula or equation will be considered as falling within the "mathematical concepts" grouping. In addition, there are instances where a formula or equation is written in text format that should also be considered as falling within this grouping. For example, the phrase "determining a ratio of A to B" is merely using a textual replacement for the particular equation (ratio = A/B). Additionally, the phrase "calculating the force of the object by multiplying its mass by its acceleration" is using a textual replacement for the particular equation (F= ma). ” The terms “modeling, Hawkes process, conditional density function, base intensity component, and cross-activation matrix components” are formulas and equation written out in text form. Therefore this claim recites an abstract idea of a mathematical concept.
defining, by the at least one processor based on the conditional density function, a log-likelihood function that is dimensionally separable; and
As stated a conditional density function is a mathematic function. A log-likelihood function is also a math function. The MPEP 2106.04(a)(2)(I)(B) states “A claim that recites a numerical formula or equation will be considered as falling within the "mathematical concepts" grouping. In addition, there are instances where a formula or equation is written in text format that should also be considered as falling within this grouping. For example, the phrase "determining a ratio of A to B" is merely using a textual replacement for the particular equation (ratio = A/B). Additionally, the phrase "calculating the force of the object by multiplying its mass by its acceleration" is using a textual replacement for the particular equation (F= ma). ” The term, “define” is a textual replacement for the mathematic transformation of a conditional density function into a log-likelihood function. Therefore this claim recites an abstract idea of a mathematical concept.
applying, by the at least one processor, a Frank-Wolfe algorithm for determining a maximum-likelihood estimation of a solution to the log-likelihood function along at least one dimension.
The Frank-Wolfe algorithm is a mathematic optimization algorithm. The maximum likelihood estimation to the log-likelihood function along at least one dimension is the mathematic optimization of the log-likelihood math function. The MPEP 2106.04(a)(2)(I)(B) states “A claim that recites a numerical formula or equation will be considered as falling within the "mathematical concepts" grouping. In addition, there are instances where a formula or equation is written in text format that should also be considered as falling within this grouping. For example, the phrase "determining a ratio of A to B" is merely using a textual replacement for the particular equation (ratio = A/B). Additionally, the phrase "calculating the force of the object by multiplying its mass by its acceleration" is using a textual replacement for the particular equation (F= ma). ” The term, “applying” is a textual replacement for the mathematic calculation of the Frank-Wolfe algorithm applied to maximize an equation. Therefore this claim recites an abstract idea of a mathematical concept.
Step 2A Prong 2, Does the claim recite additional elements that integrate the judicial exception into a practical application?
NO. Claim 1 additionally recites A method for generating sequential events data
This claim limitation states that the purpose of the claim is to generate sequential events data. The claim does not outline specifically how the abstract steps of “modeling … defining … and applying” contribute to the generation of sequential events data. The claim limitation similarly does not restrict the way that the sequential events data is generated. The MPEP 2106.05(f)(1) states “The recitation of claim limitations that attempt to cover any solution to an identified problem with no restriction on how the result is accomplished and no description of the mechanism for accomplishing the result, does not integrate a judicial exception into a practical application or provide significantly more because this type of recitation is equivalent to the words ‘apply it’” Therefore this claim does not integrate the judicial exception into a practical application or provide significantly more.
, the method being implemented by at least one processor, the method comprising:
The MPEP 2106.05(f)(2) states “Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.” Therefore the additional of at least one processor throughout the claim as a generic computing component does not integrate the judicial exception into a practical application or provide significantly more.
receiving, by the at least one processor, first information that relates to a sequence of events;
The MPEP 2106.05(f)(2) states “Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.” This claim limitation pertains to the use of generic computing components in an ordinary capacity to receive data. Therefore this limitation does not integrate the judicial exception into a practical application or provide significantly more.
Step 2B, does the claim recites additional elements that amount to significantly more than the judicial exception.
NO. As stated in Step 2A Prong 2, the above limitations do not integrate the judicial exception into a practical application or provide significantly more.
Based on the above facts, the office concludes that claim 1 is not eligible under 35 USC 101.
Claim 2:The method of claim 1, wherein the determining comprises adapting the Frank-Wolfe algorithm by adding one from among a toward-step computation and an away-step computation and applying the adapted Frank-Wolfe algorithm for determining the maximum-likelihood estimation of the solution to the log-likelihood function.
The determination step of claim 1 was determined to be an abstract idea of a mathematical concept. Adapting the mathematic Frank-Wolfe algorithm by adding one from among a toward step computation and away step computation is a modification of the mathematic algorithm through additional computations. Applying the adapted algorithm for determining the maximum-likelihood estimation is using the newly adapted mathematic formula to solve the optimization problem of claim 1. Both of these limitations are further recitations of the abstract idea of a mathematic concept introduced in claim 1 and therefore the claim pertains to an abstract idea.
Claim 3:
The method of claim 2, wherein the away-step computation comprises computing a step size by using an exact line search technique.
The term computing a step size is a textual replacement for a mathematic calculation performed during the “adapting” step outlined as a mental process in claim 2, and therefore this claim is a further recitation of the abstract ideas introduced in claim 1 and 2.
Claim 4:
The method of claim 2, wherein the away-step computation comprises computing a step size by using an adaptive step size technique.
The term computing a step size is a textual replacement for a mathematic calculation performed during the “adapting” step outlined as a mental process in claim 2, and therefore this claim is a further recitation of the abstract ideas introduced in claim 1 and 2.
Claim 5:
The method of claim 2, further comprising using a result of the applying of the adapted Frank-Wolfe algorithm to identify a sparsity pattern of the cross-activation matrix component.
The sparsity pattern of the cross -activation matrix component is a pattern of zeros in the matrix. This is observing data (the result of the applying of the adapted Frank-Wolfe algorithm) to make a evaluation/observation (identifying a sparsity pattern of the cross activation matrix component). The MPEP 2106.04(a)(2)(III) states “Accordingly, the "mental processes" abstract idea grouping is defined as concepts performed in the human mind, and examples of mental processes include observations, evaluations, judgments, and opinions. “ Therefore this claim recites a mental process.
Claim 6:
The method of claim 5, further comprising using the identified sparsity pattern to increase a convergence rate with respect to the determining of the maximum- likelihood estimation of the solution to the log-likelihood function along the at least one dimension.
Increasing a convergence rate is changing a mathematic parameter. This is done in respect to the identified sparsity pattern. This is an evaluation of the sparsity pattern and then changing a mathematic variable to solve the abstract idea of determining the MLE solution. Therefore this claim is a further recitation of the abstract idea.
Claim 7:
The method of claim 1, wherein a number of event types included in the sequence of events is equal to a number of dimensions of the multivariate Hawkes process.
This claim limitation outlines the Hawkes process which was determined to be an abstract idea of a mathematical concept. This claim changes the mathematic equations to relate to the number of event types. This is stating how the math Hawkes process is defined. Therefore this is a further recitation of the abstract idea outlined above and the claim recites an abstract idea.
Claim 8:
The method of claim 1, wherein the sequence of events is an asynchronous sequence of events for which a time interval between consecutive events is variable.
This claim expands on the type of data which is received in claim 1. This limitation is a further recitation of the use of a regular computer component to receive data which does not integrate the judicial exception into a practical application or provide significantly more than the judicial exception.
Claim 9:
The method of claim 1, wherein the sequence of events comprises at least one from among a sequence of banking events that relates to customer interactions with a financial institution, a sequence of finance events that relates to buy orders and sell orders for a particular security, a sequence of epidemiological events that relates to a spread pattern of a particular infectious disease, a sequence of advertising events that relates to click-stream data, a sequence of seismological events that relates to earthquake magnitude logs for a particular geographical region, a sequence of social media events that relates to postings for a particular social media platform, and a sequence of crime events that relates to occurrences of criminal activity in a particular neighborhood.
This claim expands on the type of data which is received in claim 1. This limitation is a further recitation of the use of a regular computer component to receive data which does not integrate the judicial exception into a practical application or provide significantly more than the judicial exception.
Claims 10-18:
Claims 10-18 are effective duplicates of claims 1-9, except that claim 10 pertain to a machine rather than a process. Furthermore, the additional limitations of A computing apparatus for generating sequential events data, the computing apparatus comprising:
a processor;
a memory;
and a communication interface coupled to each of the processor and the memory, wherein the processor is configured to:
Are generic computing machinery to implement the abstract idea. The MPEP 2106.05(f)(2) states “Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.” Therefore claims 10-18 are also direct to the judicial exception of an abstract idea which does not integrate a judicial exception into a practical application or provide significantly more
Claims 19-20:
Claims 19-20 are effective duplicates of claims 1-2, except that they pertain to a product of manufacture rather than a method. Additionally the limitations of A non-transitory computer readable storage medium storing instructions for generating sequential events data, the storage medium comprising executable code which, when executed by a processor, causes the processor to:
Are generic computing machinery to implement the abstract idea. The MPEP 2106.05(f)(2) states “Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.” Therefore claims 19-20 are also direct to the judicial exception of an abstract idea which does not integrate a judicial exception into a practical application or provide significantly more.
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.
Claims 1, 10, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Ide_2022 (US 20220391736 A1) and Kent_2021 (Modified Frank Wolfe in Probability Space)
Claim 1:
Ide_2022 makes obvious A method for generating sequential events data (par 3: “Modeling time-stamped events using point processes is an emerging research topic in machine learning (ML) gaining considerable recent interest.”) , the method being implemented by at least one processor, the method comprising: (par 112: “ The present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.”)
receiving, by the at least one processor, first information that relates to a sequence of events; (par 6: “The computer-implemented method includes receiving an event log, where the event log includes timestamps and event types.”)
modeling, by the at least one processor based on the received first information, the sequence of events by a multidimensional Hawkes process (par 37: “Embodiments of the present invention propose a unified approach to a problem, in which not only is causal relationships between the various event types learned but also causal association probabilities between individual event instances are determined. For the former, embodiments of the present invention develop a cardinality regularization technique in fitting multivariate Hawkes processes. This achieves causal estimation that is both accurate and sparse thus helping to enable effective event consolidation.”) that relates to a conditional density function (par 51: “Given custom-character.sub.n-1, the intensity function is defined as the probability density that the first event since t.sub.n-1 occurs. This is a conditional density. Examiners note: this is equation (6)) … par 56: “Equations (6) and (9) hold for any point process. Here, we introduce a specific parameterization of the Hawkes process:” that includes a base intensity component and a cross- activation matrix component; (par 56: “where μ.sub.d≥0 is called the baseline intensity of the d-th type, A.sub.d,d.sub.i is the (d, d.sub.i)-element of the impact matrix A∈”)
defining, by the at least one processor based on the conditional density function, a log-likelihood function that is dimensionally separable; and (par 52: “Integrating the both sides of equation (6) and arranging the terms, we obtain (examiner note: equation (8)) … which allows representing L.sub.0 in terms of the intensity). Examiner note: where L is the likelihood function, see par 47: “This decomposition readily leads to the definition of the base likelihood function L.sub.0:” and the intensity referenced is from equation (6), which is the conditional density as shown above.
applying, by the at least one processor, par 60: “ Numerically solving for maximum likelihood estimation (MLE) is known to be challenging even when τ=0, mainly due to the nonlinear logarithmic term in equation (9). The minorization-maximization (MM) algorithm leverages the additive structure of the Hawkes process in equation (10) to apply Jensen's inequality in a manner similar to the expectation-maximization (EM) algorithm for mixture models (Neal et al., 1998). Specifically, we first rewrite equation (10) ”)
Does not expressly recite a Frank-Wolfe
Kent_2021 however makes obvious Frank-Wolfe (4 Computational examples page 7 par 5: “In this section, we demonstrate the application our Frank-Wolfe algorithm to several non-parametric estimation problems in statistics and machine learning.)
Ide_2022 and Kent_2021 are analogous art to the claimed invention because they are from the same field of endeavor called machine learning/equation optimization. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Ide_2022 and Kent_2021.
The rationale for doing so would have been to follow a teaching proposed in the art. Ide_2022 teaches using a minorization-maximation algorithm to determine a MLE to a log-likelihood function. Kent_2021 teaches the use of a frank-Wolfe algorithm to solve for the optima of a MLE.
PNG
media_image1.png
372
936
media_image1.png
Greyscale
PNG
media_image2.png
767
861
media_image2.png
Greyscale
Therefore, it would have been obvious to combine the workflow to solve for an optima of the Hawkes process of Ide_2022 with the usage of a Frank-Wolfe algorithm of Kent_2021 for the benefit of achieving a quick global optima to obtain the invention as specified in the claims.
Claim 10:
Claim 10 is effectively similar to claim 1 and is rejected under the same rational. Additionally Ide_2022 makes obvious the additional limitations of A computing apparatus for generating sequential events data, the computing apparatus comprising:
a processor;
a memory;
and a communication interface coupled to each of the processor and the memory, wherein the processor is configured to: (par 117: “These computer readable program instructions may be provided to a processor of a computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.”)
Claim 19:
Claim 19 is effectively similar to claim 1 and is rejected under the same rational. Additionally Ide_2022 makes obvious the additional limitations of A non-transitory computer readable storage medium storing instructions for generating sequential events data, the storage medium comprising executable code which, when executed by a processor, causes the processor to: (par 113: “The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.”)
Claims 2-5, 11-14, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Ide_2022 Kent_2021, and Bomze_2021 (“Frank–Wolfe and friends: a journey into projection-free first-order optimization methods”)
Claim 2:The method of claim 1, wherein the determining comprises
Ide_2022 makes obvious see claim 1)
Ide_2022 does not expressly recite adapting the Frank-Wolfe algorithm by adding one from among a toward-step computation and an away-step computation and applying the adapted Frank-Wolfe algorithm for
Bomze_2021 however makes obvious adapting the Frank-Wolfe algorithm by adding one from among a toward-step computation and an away-step computation and applying the adapted Frank-Wolfe algorithm for (page 314 par 3: “The slow convergence behaviour for objectives with solution on the boundary motivated the introduction of several variants, the most popular being Wolfe’s away step (Wolfe 1970). Wolfe’s idea was to move away from bad vertices, in case a step of the FW method moving towards good vertices did not lead to sufficient improvement on the objective. This idea was successfully applied in several network equilibrium problems, where linear minimization can be achieved by solving a min-cost flow problem (see Fukushima 1984 and references therein)”
Ide_2022, Kent_2021 and Bomze_2021 are analogous art to the claimed invention because they are from the same field of endeavor called machine learning/ equation optimization. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Ide_2022, Kent_2021 and Bomze_2021. The rationale for doing so would have been to follow a teaching and motivation proposed in the prior art. Bonze_2021 states that the away step is a popular variant used to solve objectives with boundary solutions. Therefore, it would have been obvious to combine the workflows of Ide_2022 and Kent_2021 that solves for a MLE with Frank-Wolfe algorithm with the teaching of an away step of Bomze_2021 for the predictable and popular benefit of solving for boundary cases to obtain the invention as specified in the claims.
Claim 3:
The method of claim 2,
Ide_2022 does not expressly recite wherein the away-step computation comprises computing a step size by using an exact line search technique.
Bomze_2021 however makes obvious wherein the away-step computation (page 314 par 3: “The slow convergence behavior for objectives with solution on the boundary motivated the introduction of several variants, the most popular being Wolfe’s away step) comprises computing a step size by using an exact line search technique. (page 324 4 Stepsizes: “Popular rules for determining the stepsize are: … exact line search”) Examiner note: Where both using an away step being popular, and determining a stepsize using exact line search being popular, makes obvious the usage of an exact learn search the compute the stepsize of an away step.
As stated earlier, it would have been obvious to combine the workflows of Ide_2022 and Kent_2021 that solves for a MLE with Frank-Wolfe algorithm with the teaching of an away step and the exact line search technique of Bomze_2021 for the predictable and popular benefit of solving for boundary cases to obtain the invention as specified in the claims.
Claim 4:The method of claim 2,
Ide_2022 does not expressly recite wherein the away-step computation comprises computing a step size by using an adaptive step size technique.
Bomze_2021 however makes obvious wherein the away-step computation (page 314 par 3: “The slow convergence behavior for objectives with solution on the boundary motivated the introduction of several variants, the most popular being Wolfe’s away step) comprises computing a step size by using an adaptive step size technique. (page 324 4 Stepsizes: “Popular rules for determining the stepsize are: … diminishing stepsize”) Examiner note: Where the examiner interprets a diminishing stepsize to be an adaptive step size technique. Where the passage makes obvious using an away step being popular, and determining a stepsize using adaptive step sizes being popular, makes obvious the usage of an adaptive step size for an away step.
As stated earlier, it would have been obvious to combine the workflows of Ide_2022 and Kent_2021 that solves for a MLE with Frank-Wolfe algorithm with the teaching of an away step and the exact line search technique of Bomze_2021 for the predictable and popular benefit of solving for boundary cases to obtain the invention as specified in the claims.
Claim 5:
The method of claim 2,
Ide_2022 makes obvious further comprising using a result of the applying of the [minorization-maximization (MM)] (Par 33: “ Table 1 summarizes L.sub.0Hawkes, the proposed algorithm, which is used as part of the iterative MM procedure in equation (24). “ … Par 96: “At step 405, the computing device or server computes a sparse impact matrix (A) with cardinality regularization, using the initial triggering probabilities {q.sub.n,i}. Based on the initial triggering probabilities {q.sub.n,i}, the sparse impact matrix (A) is computed by using the algorithm presented in Table 1.”)
Ide_2022 does not expressly recite adapted Frank-Wolfe algorithm
Bromze_2021 however makes obvious adapted Frank-Wolfe algorithm (page 317 par 2: “FW (Examiner note: Frank-Wolfe) methods and variants (examiner note: adapted) are a natural choice for constrained optimization on convex sets admitting a linear minimization oracle significantly faster than computing a projection. We present here in particular the traffic assignment problem, submodular optimization, LASSO problem, matrix completion, adversarial attacks, minimum enclosing ball, SVM training, maximal clique search in graphs, sparse optimization”)
Ide_2022 and Bomze_2021 are analogous art to the claimed invention because they are from the same field of endeavor called machine learning/ function optimization. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Ide_2022 and Bomze_2021.
The rationale for doing so would have been to follow a teaching in the prior art. Ide_2022 teaches a workflow which computes a sparse matrix pattern. The prior art of Ide_2022 does not use an adapted Frank-Wolfe algorithm, but instead relies on the Table 1 minorization maximization algorithm. The prior art of Bomze_2021 states page 314 par 5: “One of the main features of the FW algorithm is its ability to naturally identify sparse and structured (approximate) solutions. For instance, if the optimization domain is the simplex, then after k steps the cardinality of the support of the last iterate generated by the method is at most k + 1. Most importantly, in this setting every vertex added to the support at every iteration must be the best possible in some sense, a property that connects the method with many greedy optimization schemes(Clarkson 2010). This makes the FW method pretty efficient on the abovementioned problem class. Indeed, the combination of structured solutions with often noisy data makes the sparse approximations found by the method possibly more desirable than high precision solutions generated by a faster converging approach.
Therefore, it would have been obvious to combine the workflow for time series modeling with sparsity matrix of Ide_2022 with the usage of an adapted Frank-Wolfe algorithm of Bomze_2021 for the benefit of gaining efficient solutions to naturally identify sparse solutions to obtain the invention as specified in the claims.
Claims 11-14:
Claims 11-14 are effectively similar to claims 2-5 except that they depend from claim 10 and are therefore rejected under the same rational as claims 2-5 and 10.
Claim 20:
Claim 20 is effectively similar to claim 2 except that it depends from claim 19 and is therefore rejected under the same rational as claims 2 and 19,
Claims 6 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Ide_2022 Kent_2021, Bomze_2021, and Ziganto_2017 (“ Sparse Matrices For Efficient Machine Learning”)
Claim 6:
The method of claim 5, further comprising
Ide_2022 makes obvious using the identified sparsity pattern (par 105: “FIG. 5(A) shows a sparsity pattern of estimated impact matrix A with L.sub.0Hawkes, FIG. 5(B) shows a sparsity pattern of estimated impact matrix A with the l.sub.1-regularizer, and FIG. 5(C) shows a sparsity pattern of estimated impact matrix A with the l.sub.2,1-regularizer.”) along the at least one dimension. (par 59: “To guarantee sparsity, we propose the following cardinality-regularized maximum likelihood:”)Examiner note: see also figure 4.
PNG
media_image3.png
746
679
media_image3.png
Greyscale
Ide_2022 does not expressly recite that the use of a sparsity matrix leads to increase a convergence rate
Ziganto_2017 however makes obvious to increase a convergence rate (introduction: “Because sparse matrices have lots of zero values, we can apply special algorithms that will do two important things:
compress the memory footprint of our matrix object
speed up many machine learning routines
“)
Ide_2022 and Ziganto_2017 are analogous art to the claimed invention because they are from the same field of endeavor called machine learning/function optimization. Before the effective filing date, it would have been obvious to a person ordinarily skilled in the art to combine Ide_2022 and Ziganto_2017. The rational for doing so would have been to follow a teaching and motivation proposed in the art. The prior art of Ide_2022 teaches sparsity matrixes and incorporating it into the MLE, see par 59: “To guarantee sparsity, we propose the following cardinality-regularized maximum likelihood:”) Ziganto_2017 makes obvious that sparsity can cause an increase in a convergence rate alongside algorithms. One ordinarily skilled in the art would recognize that the existing process taught by Ide_2022 which uses a sparsity matrix will cause an increase the convergence rate due to the decrease in data. Therefore it would have been obvious to combine the workflow of Ide_2022 with sparsity matrix alongside algorithms which allow sparsity matrix can be used to speed up machine learning routines of Ziganto_2017 for the benefit of running the workflow of Id_2022 with faster algorithms to obtain the invention as specified in the claims.
Claim 15:
Claim 15 is effectively similar to claim 6, except that it depends from claim 14, and is therefore rejected under the same rationales as claims 6 and 14.
Claims 7-9 and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Ide_2022 Kent_2021, and Remi_2017 (“Multivariate Hawkes Processes for Large-Scale Inference”)
Claim 7:
The method of claim 1,
Ide_2022 does not expressly recite wherein a number of event types included in the sequence of events is equal to a number of dimensions of the multivariate Hawkes process.
Remi_2017 however makes obvious wherein a number of event types included in the sequence of events is equal to a number of dimensions of the multivariate Hawkes process. (abstract: “In this paper, we present a framework for fitting multivariate Hawkes processes for large-scale problems, both in the number of events in the observed history n and the number of event types d (i.e. dimensions).”)
Ide_2022 and Remi_2017 are analogous art to the claimed invention because they are from the same field of endeavor called machine learning and function optimization. Before the effective filing date, it would have been obvious to a person ordinarily skilled in the art to combine Ide_2022 and Remi_2017.
The rational for doing so would have been Applying a known technique to a known device ready for improvement to yield a predictable result. Ide_2022 teaches a known technique (the usage of multiple event types in a multivariate Hawkes process) See par 37: “embodiments of the present invention develop a cardinality regularization technique in fitting multivariate Hawkes processes.” And abstract : “The computer determines a sparse impact matrix representing causal relationships between the event types”) Ide_2022 does not state that the dimensions of the model are based on the number of event types. Remi_2017 implies that event types is understood by one ordinarily skilled in the art to mean a number of dimensions in the model.
Therefore it would have been obvious to combine the usage of a multivariate Hawkes process of different event types of Ide_2022 with a dimensions of each event type of Remi_2017 to apply the known technique of the multivariate Hawkes process which contains a dimension for each event type to obtain the invention as specified in the claims.
Claim 8:
The method of claim 1,
Ide_2022 does not expressly recite wherein the sequence of events is an asynchronous sequence of events for which a time interval between consecutive events is variable.
Remi_2017 however makes obvious wherein the sequence of events is an asynchronous sequence of events for which a time interval between consecutive events is variable. (page 2168 introduction par 1-3: “ Multivariate Hawkes processes (MHP) (Oakes 1975; Liniger 2009) have emerged in several fields as the gold standard to deal with such data, e.g. earthquake prediction (Vere Jones 1978), biology (Reynaud-Bouret et al. 2014), financial (Bauwens and Hautsch 2009; Alfonsi and Blanc 2015), and social interactions studies (Crane and Sornette 2008). MHP, an event of type u (e.g. a visit to a product’s web site) occurring at time t, will increase the conditional occurrence rate of events of type v at time s≥t (e.g. purchases of that product in the future) by a rate guv(s−t).”
Ide_2022 and Remi_2017 are analogous art to the claimed invention because they are from the same field of endeavor called machine learning and function optimization. Before the effective filing date, it would have been obvious to a person ordinarily skilled in the art to combine Ide_2022 and Remi_2017.
The rational for doing so would have been to follow a teaching proposed in the art. One normally skilled in the art could likely infer and understand that the prior art of Ide_2022 which uses a multivariate Hawkes process deals with an asynchronous sequence of events. See par 106: “Par 106: “ In the second real use-case, we applied L.sub.0Hawkes to a real event triage task. We obtained N=718 warning events from a real cloud data center management system. These events resulted from filtering logs emitted by network devices and each has its type: there were D=14 unique event types in our dataset. In this real use-case, we focused on showing examples of instance-level causal analysis.“ However Ide_2022 does not expressly state this. Remi_2017 does expressly state this, and states that in a MHP (multivariate Hawkes process) an occurrence of an event changes the occurrence rate of other events. One ordinarily skilled in the art understands that the MHP definition of using variable data applies to the prior art of Ide_2022, since warning events do not occur synchronously, and it would have been obvious to do so to follow the Remi_2017 introduction par 2: “gold standard to deal with such data,”
Therefore it would have been obvious to combine the multivariate Hawkes process with data of Ide_2022 with the express definition of Remi_2017 which states that MHP data is asynchronous for the benefit of processing asynchronous data, (such as a cloud center management system) to obtain the invention as specified in the claims.
Claim 9:The method of claim 1, wherein the sequence of events comprises
Ide_2022 does not expressly recite at least one from among a sequence of banking events that relates to customer interactions with a financial institution, a sequence of finance events that relates to buy orders and sell orders for a particular security, a sequence of epidemiological events that relates to a spread pattern of a particular infectious disease, a sequence of advertising events that relates to click-stream data, a sequence of seismological events that relates to earthquake magnitude logs for a particular geographical region, a sequence of social media events that relates to postings for a particular social media platform, and a sequence of crime events that relates to occurrences of criminal activity in a particular neighborhood.
Remi_2017 however makes obvious at least one from among a sequence of banking events that relates to customer interactions with a financial institution, a sequence of finance events that relates to buy orders and sell orders for a particular security, a sequence of epidemiological events that relates to a spread pattern of a particular infectious disease, a sequence of advertising events that relates to click-stream data, a sequence of seismological events that relates to earthquake magnitude logs for a particular geographical region, a sequence of social media events that relates to postings for a particular social media platform, and a sequence of crime events that relates to occurrences of criminal activity in a particular neighborhood. (page 2168 introduction par 1: “In finance, arrivals of buying and selling orders for different stocks convey information about macroscopic market tendencies.” … page 2168 introduction par 2: “Multivariate Hawkes processes (MHP) (Oakes 1975; Liniger 2009) have emerged in several fields as the gold standard to deal with such data)
it would have been obvious to combine the multivariate Hawkes process with data of Ide_2022 with the usage of financial data of Remi_2017 for the benefit of utilizing the gold standard to process such data to identify market tendencies to obtain the invention as specified in the claims.
Claims 16-18:Claims 16-18 are effectively similar to claims 7-9 except that they depend from claim 10 and are therefore rejected under the same rational as claims 7-9 and 10.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Okawa_2021 (US 20210209496 A1) “SPATIO-TEMPORAL EVENT DATA ESTIMATING DEVICE, METHOD, AND PROGRAM” has a multidimensional spatio-temporal Hawkes process and estimates parameters to optimize a likelihood function for events occurring.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AHMAD HUSSAM SHALABY whose telephone number is (571)272-7414. The examiner can normally be reached Mon-Fri 7:30am - 5pm.
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, Emerson Puente can be reached at 5712723652. 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.
/A.H.S./Examiner, Art Unit 2187
/EMERSON C PUENTE/Supervisory Patent Examiner, Art Unit 2187
Prosecution Insights