DETAILED ACTION
This office action is responsive to the response filed 6/2/2026. The application contains claims 1-6, 9-15, 18-20, all examined and rejected.
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 .
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/process/file/efs/guidance/eTD-info-I.jsp.
Claims 1-6, 9-15, and 18-20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1-16 of U.S. Patent No. US 10,535,012 B2. Although the claims at issue are not identical, they are not patentably distinct from each other because each of the instant claims (the claim being examined) is “generic to a species or sub-genus claimed in a conflicting patent or application, i.e., the entire scope of the reference claim falls within the scope of the examined claim.” See MPEP 804(II)(B)(1).
Instant Application
U.S. Patent No. US 10,535,012 B2 (reference patent)
Claim 1
An apparatus for implementing a computing system to predict preferences, the apparatus comprising: at least one processor device operatively coupled to a memory and configured to:
generate a plurality of samples from a prior distribution;
obtain, from a behavioral model of at least one person, a likelihood of observation for each sample of the plurality of samples, wherein the behavioral model has an internal state estimated by the prior distribution;
eliminate a first set of samples from the plurality of samples to generate a second set of samples from the plurality of samples, wherein the likelihood of observation of each sample of the first set of samples is less than a threshold value, and the second set of samples is different from the first set of samples;
calculate a parameter relating to a density of a prior distribution at each sample of a set of samples associated with the prior distribution; and
estimate, for the plurality of samples, at least one differential entropy of at least one posterior distribution associated with at least one observation based on the parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample,
select an action from a plurality of candidate actions, wherein each candidate action of the plurality of candidate actions cause the at least one observation; and transmit, to at least one device associated with the at least one person, at least one electronic interaction generated based on the action.
Claim 1
An apparatus to improve operation of a computing system for predicting personal preferences, comprising: a processor operatively coupled to a memory and configured to:
generate a plurality of samples from a prior distribution, the prior distribution including a distribution of values representing at least one preference of at least one person;
obtain, for each sample among the plurality of samples, a likelihood of an observation as an output of a likelihood function given the sample;
eliminate samples from the plurality of samples having a likelihood of observation less than a threshold value to generate a subset of the plurality of samples;
calculate at least one parameter relating to a density of the prior distribution at each sample in the subset, the at least one parameter including a distance from each sample to at least one neighboring sample;
estimate, for each sample in the subset, at least one differential entropy of at least one posterior distribution associated with at least one observation based on the at least one parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample, the estimation being performed on samples in the subset; and
transmit, to at least one device associated with the at least one person, at least one electronic interaction generated based on an action, the action being selected based on expected values of the differential entropies estimated for all observations caused by the action.
Claim 2
The apparatus of claim 1,
wherein the at least one processor device is further configured to:
obtain, for each sample of the second set of samples, the likelihood of observation as an output of a likelihood function given a respective sample of the second set of samples.
Claim 1
An apparatus to improve operation of a computing system for predicting personal preferences, comprising: a processor operatively coupled to a memory and configured to:
generate a plurality of samples from a prior distribution, the prior distribution including a distribution of values representing at least one preference of at least one person;
obtain, for each sample among the plurality of samples, a likelihood of an observation as an output of a likelihood function given the sample;
eliminate samples from the plurality of samples having a likelihood of observation less than a threshold value to generate a subset of the plurality of samples;
calculate at least one parameter relating to a density of the prior distribution at each sample in the subset, the at least one parameter including a distance from each sample to at least one neighboring sample;
estimate, for each sample in the subset, at least one differential entropy of at least one posterior distribution associated with at least one observation based on the at least one parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample, the estimation being performed on samples in the subset and without sampling the at least one posterior distribution to reduce consumption of resources of the computing system; and
transmit, to at least one device associated with the at least one person, at least one electronic interaction generated based on an action, the action being selected based on expected values of the differential entropies estimated for all observations caused by the action.
Claim 3
The apparatus of claim 1, wherein a distance from each sample of the second set of samples to at least one neighboring sample of the second set of samples is a distance from each sample of the second set of samples to a kth-nearest neighbor of the second set of samples, k being a natural number.
Claim 2
The apparatus of claim 1, wherein the processor is further configured to calculate a distance from each sample to a kth-nearest neighbor as the at least one parameter relating to the density at each sample, k being a natural number.
Claim 4
The apparatus of claim 1, wherein the at least one processor device is further configured to estimate the at least one differential entropy of the at least one posterior distribution by approximating a probability density function of the prior distribution at each sample of the second set of samples using a volume of a sphere having a radius equal to the distance.
Claim 3
The apparatus of claim 1, wherein the processor is further configured to estimate the at least one differential entropy of the at least one posterior distribution by approximating a probability density function of the prior distribution at each sample using a volume of a sphere having a radius equal to the distance.
Claim 5
The apparatus of claim 1, wherein the at least one processor device is further configured to estimate, using Euler's constant as a constant term the at least one differential entropy of the at least one posterior distribution.
Claim 4
The apparatus of claim 1, wherein the processor is further configured to estimate the at least one differential entropy of the at least one posterior distribution having Euler's constant as a constant term.
Claim 6
The apparatus of claim 1,
wherein the at least one processor device is further configured to estimate the at least one differential entropy of each posterior distribution of a plurality of posterior distributions based on the parameter relating to the density of the prior distribution at each sample of the second subset of samples and a likelihood of transition for each sample of the second set of samples from the prior distribution to each posterior distribution of the plurality of posterior distribution, and
wherein the likelihood of transition of each sample of the second set of samples exceeds a threshold likelihood and the plurality of posterior distributions includes the at least one posterior distribution.
Claim 1
An apparatus to improve operation of a computing system for predicting personal preferences, comprising: a processor operatively coupled to a memory and configured to:
estimate, for each sample in the subset, at least one differential entropy of at least one posterior distribution associated with at least one observation based on the at least one parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample, the estimation being performed on samples in the subset and without sampling the at least one posterior distribution to reduce consumption of resources of the computing system;
Claim 6
The apparatus of claim 1, wherein the subset of the plurality of samples are samples having a likelihood of transition exceeding a threshold likelihood.
Claim 9
The apparatus of claim 1, wherein the action is selected based on expected values of the at least one differential entropy estimated for the at least one observation caused by the action
Claim 1
“...transmit, to at least one device associated with the at least one person, at least one electronic interaction generated based on an action, the action being selected based on expected values of the differential entropies estimated for all observations caused by the action.”
Claim 20
The computer program product of claim 19, wherein the action from the plurality of candidate actions each causing one or more observations is selected based on expected values of the differential entropies estimated for all observations caused by the action.
Claim 16
A computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform operations for improving operation of a computing system for predicting personal preferences, the operations comprising:
transmitting, to at least one device associated with the at least one person, at least one electronic interaction generated based on an action, the action being selected based on expected values of the differential entropies estimated for all observations caused by the action.
As indicated in the table above, all the claimed features in instant claim 1 are disclosed in reference claim 1. While the two claims are not identical, instant claim 1 is anticipated by reference claim 1. It is evident from the table that all limitations in instant claim 1 are linguistically comparable to the underlined limitations in reference claim 1 except for the limitation “calculate a parameter relating to a density of a prior distribution at each sample of a set of samples associated with the prior distribution” in instant claim 1, for which explanation is provided below:
Reference claim 1 recites “calculate at least one parameter relating to a density of the prior distribution at each sample in the subset” wherein the “subset” refers to “samples from the plurality of samples having a likelihood of observation less than a threshold value” and the plurality of samples are associated with the prior distribution (see the “generate...” and “obtain...” limitations of reference claim 1). Therefore, reference claim 1 anticipates instant claim 1.
Instant claims 10 and 19 recite analogous limitations as claim 1, and are rejected based on similar rationale as stated above for claim 1 (instant claim 10 compared to reference claim 13; instant claim 19 compared to reference claim 16). Each of the instant dependent claims as noted above is rejected based on the same rationale as the claim from which it depends. Please see table for more information.
Regarding claim 10;
Claim 10 is similar in scope to claim 1; therefore, it is rejected under similar rationale.
Regarding claim 11;
Claim 11 is similar in scope to claim 2; therefore, it is rejected under similar rationale.
Regarding claim 12;
Claim 12 is similar in scope to claim 3; therefore, it is rejected under similar rationale.
Regarding claim 13;
Claim 13 is similar in scope to claim 4; therefore, it is rejected under similar rationale.
Regarding claim 14;
Claim 14 is similar in scope to claim 5; therefore, it is rejected under similar rationale.
Regarding claim 15;
Claim 15 is similar in scope to claim 6; therefore, it is rejected under similar rationale.
Regarding claim 18;
Claim 18 is similar in scope to claim 9; therefore, it is rejected under similar rationale.
Regarding claim 19;
Claim 19 is similar in scope to claim 1; therefore, it is rejected under similar rationale. Further U.S. Patent No. 10,535,012 B2 teach A computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform operations for improving operation of a computing system for predicting personal preferences (Claim 16).
Regarding claim 20;
Claim 20 is similar in scope to claim 9; therefore, it is rejected under similar rationale.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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 1-3, 9-12, and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Balakrishnan et al. (US 2012/0143802 A1, hereinafter Balakrishnan) in view of Pant et al. (“An information-theoretic approach to assess practical identifiability of parametric dynamical systems”, hereinafter Pant) Disclosed in IDS submitted 9/25/2025 and further in view of “A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking” Published 2002 [hereinafter D1].
Regarding Claim 1,
Balakrishnan teaches an apparatus for implementing a computing system to predict preferences, the apparatus comprising:
at least one processor device operatively coupled to a memory and configured to (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences),
from a prior distribution, obtain, from a behavioral model of at least one person (¶24, “That is, the prior on u★ is: … (Equation #2)”, ¶22, “assume a multivariate Normal form for the prior distribution of the current user parameter vector”, ¶16, “unobserved user-specific and item-specific latent (or hidden) parameters 30 whose combination determines a preference 32 that the user will have for an item”, ¶39, “assume a MVN prior distribution of the user parameter vector: p(u)= (u|μ 0,Σ0)“, ¶16, “Latent factor models may generally assume that there are unobserved user-specific and item-specific latent (or hidden) parameters 30 whose combination determines a preference 32 that the user will have for an item”), a likelihood of observation, wherein the behavioral model has an internal state estimated by the prior distribution (¶25, Equation #5, “the model allows the inventors to compute the likelihood of any particular parameter value for a user u★, given the pair parameters θlr and the user preference 32 over these items yl>r”, ¶39, “t z(u)=P(y z|θZ ,u) is the likelihood term for the zth feedback pair“, ¶30, “the user parameters are MVN distributed: p(u)= (u|μ N,ΣN)“, ¶18, “Each successive response 40, though, may be incorporated into the latent factor model 28 to recursively refine an estimate of the user's parameters 30”);
estimate, at least one differential entropy of at least one posterior distribution associated with at least one observation (¶30, “the (differential) entropy h for a k-dimensional MVN distribution with covariance matrix S, is given by the expression: …”, “this results in the total information gain criterion, IGlr=hN+1−hN”)”, “that given any one pair of items l and r, its pair parameters θlr and pairwise binary response yl>r”, ¶29, “This criterion approximates the expected change in entropy or information gain between the user parameter distribution before and after receiving feedback, for any pair of items”);
select an action from a plurality of candidate actions, wherein each candidate action of the plurality of candidate actions causes the at least one observation (¶31, “the choice of the next pair of items (l, r) to get feedback for would be ones that maximize IGlr“, ¶37, “After applying the filters, exemplary embodiments obtain a filtered list of candidate items. Exemplary embodiments may then evaluate the IG criterion on all pairs of these candidate items”, ¶22, “find the most informative pair of items for which to obtain feedback”, ¶18); and
transmit, to at least one device associated with the at least one person, at least one electronic interaction generated based on the action (Fig. 1, ¶15, “FIG. 1 is a simplified schematic illustrating an environment in which exemplary embodiments may be implemented. FIG. 1 illustrates a client-server network architecture that recommends items to users. A server 20 communicates with a client device 22 via a communications network 24”, ¶18, “the user may be asked whether she prefers the “The Godfather” or “Annie Hall.””, ¶21, The recommender application 26 sends the one or more pairwise questions 38 to the client device 22, and the client device 22 sends the user's responses 40. The recommender application 26 may then incorporate each successive response 40 into the latent factor model 28 to refine an estimate of the user's parameters 30”).
Balakrishnan does not appear to explicitly teach calculate a parameter relating to a density of a prior distribution at each sample of a set of samples; wherein the at least one differential entropy is estimated based on the parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample of the set of samples.
However, Pant teaches calculate a parameter relating to a density of a prior distribution at each sample of a set of samples associated with the prior distribution (pg. 71 Section 6.1:
PNG
media_image1.png
848
710
media_image1.png
Greyscale
teaches calculating the distance between each sample to a kth-nearest neighbor relating to density; pg. 67 Section 2 Equation (2) teaches density of a prior distribution); and
estimate, for the set of samples, at least one differential entropy of at least one posterior distribution associated with at least one observation wherein the at least one differential entropy is estimated based on the parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample of the set of samples (pg. 67 Section 2:
PNG
media_image2.png
405
592
media_image2.png
Greyscale
PNG
media_image3.png
217
597
media_image3.png
Greyscale
P. 72, “Algorithm1: Algorithm to calculate gain in information for the parameters”, Line 2, lines 18-21, teaches estimating the differential entropy of the posterior distribution by updating the prior probability distribution to the posterior distribution using Bayes’ theorem and using the probability density function px(x), which corresponds to estimating the differential entropy of a posterior distribution associated with a parameter related to density; Equation (2) teaches that the likelihood of observation is used to calculated the posterior probability distribution and that the posterior distribution is not sampled in the calculation of differential entropy; instead, the prior distribution is sampled; the estimation of the differential entropy of the posterior distribution is performed on the subset of px|y(x|y)).
Balakrishnan and Pant are analogous art because they are directed to analysis related to posterior distributions.
It would have been obvious for one of ordinary skill in the arts before the effective filing date of the claimed invention to incorporate calculate a parameter relating to a density of a prior distribution at each sample of a set of samples associated with the prior distribution; estimate, for the plurality of samples, at least one differential entropy of at least one posterior distribution associated with at least one observation based on the parameter relating to the density of the prior distribution at each sample and the likelihood of observation for each sample as taught by Pant et al. to the disclosed invention of Balakrishnan.
One of ordinary skill in the arts would have been motivated to make this modification in order to provide a framework for quantification of information gain measurements that is easily parallelisable which allows for efficient, scalable analysis of large datasets and as analytical tractability of the probability distributions and their integration to calculate entropy is a concern, especially when the forward and observation operators are non-linear and dimensionality of the problem is high. Nonetheless, one may be able to simulate a large, yet finite, number of particles (samples from the prior distribution) forward in time to obtain samples from the probability distribution of Z1:j. For this reason, methods to estimate H(X) only from samples of X are sought (Pant pg. 66-67 Section 1, pg. 70-71 Section 6). This is simply combining prior art elements according to known methods to yield predictable results and applying a known technique to a known device (method, or product) ready for improvement to yield predictable results (MPEP 2143).
Balakrishnan-Pant does not explicitly teach generate a plurality of samples; a likelihood of observation for each sample of the plurality of samples, eliminate a first set of samples from the plurality of samples to generate a second set of samples from the plurality of samples, wherein the likelihood of observation of each sample of the first set of samples is less than a threshold value, and the second set of samples is different from the first set of samples.
However, D1 teaches generate a plurality samples from the prior distribution (P. 728, “choose the importance density to be the prior”
PNG
media_image4.png
57
504
media_image4.png
Greyscale
, P. 780, Algorithm 4,
PNG
media_image5.png
37
272
media_image5.png
Greyscale
);
obtain, for each sample of the plurality of samples, a likelihood of observation as an output of a likelihood function given a respective sample (P. 730, “For this particular choice of importance density, it is evident that the weights are given by
PNG
media_image6.png
49
434
media_image6.png
Greyscale
PNG
media_image7.png
122
680
media_image7.png
Greyscale
,
PNG
media_image8.png
454
655
media_image8.png
Greyscale
”)
eliminate a first set of samples from the plurality of samples to generate a second set of samples from the plurality of samples, wherein the likelihood of observation of each sample of the first set of samples is less than a threshold value, and the second set of samples is different from the first set of samples (P. 730, “For this particular choice of importance density, it is evident that the weights are given by
PNG
media_image6.png
49
434
media_image6.png
Greyscale
PNG
media_image7.png
122
680
media_image7.png
Greyscale
, P. 728, “use resampling whenever a significant degeneracy is observed (i.e., when Ne / f falls below some threshold NT). The basic idea of resampling is to eliminate particles that have small weights and to concentrate on particles with large weights”, P. 728, Eq. 64, P. 729, Algorithm 2, particle weight is observation given that sample).
Balakrishnan-Pant and D1 are analogous art because they are directed to the usage of likelihood function to produce posterior distributions.
It would have been obvious for one of ordinary skill in the arts before the effective filing date of the claimed invention to incorporate generate a plurality samples from the prior distribution; obtain, for each sample among the plurality of samples, a likelihood of an observation as an output of a likelihood function given the sample; and eliminate samples from the plurality of samples having a likelihood less than a threshold value to generate the set of samples as taught by D1 to the disclosed invention of Balakrishnan-Pant.
One of ordinary skill in the arts would have been motivated to make this modification in order to use a recursive filtering approach so that received data can be processed sequentially rather than as a batch so that it is not necessary to store the complete data set nor to reprocess existing data if a new measurement becomes available which save resources and energy (D1, P. 723, Col. 2, ¶3, P. 728, “degeneracy implies that a large computational effort is devoted to updating particles whose contribution to the approximation to is almost zero”). This is simply combining prior art elements according to known methods to yield predictable results and applying a known technique to a known device (method, or product) ready for improvement to yield predictable results (MPEP 2143).
Regarding Claim 2,
Balakrishnan-Pant-D1 teaches the apparatus of claim 1, wherein the at least one processor device is further configured to:
obtain, for each sample of among the second set plurality of samples, the likelihood of observation as an output of a likelihood function given a respective sample of the second set of samples (Balakrishnan, ¶25, Eq. (5), “the model allows the inventors to compute the likelihood of any particular parameter value for a user u★, given the pair parameters θlr and the user preference 32 over these items yl>r”, ¶39, “t z(u)=P(y z|θZ ,u) is the likelihood term for the zth feedback pair, which can be evaluated using the expression: t z(u)=Φ(δlr z)y l>r z (1−Φ(δlr z))(1−y l>r z )”, ¶30, “the log-likelihood of the user parameters 30 can be obtained using Equation #5 as: …“, D1, P. 729, “Algorithm 3: Generic Particle Filter”, resampling occur within the recursion, so the particles at next steps are the resamples ones, P. 730, Algorithm 4: SIR Particle Filter
PNG
media_image9.png
103
456
media_image9.png
Greyscale
).
Regarding Claim 3,
Balakrishnan-Pant-D1 teaches the apparatus of claim 1, wherein a distance from each sample of the second set of samples to at least one neighboring sample of the second set of samples is a distance from each sample of the second set of samples to a kth nearest neighbor sample of the second set of samples, k being a natural number. (Pant, P. 72, “Algorithm1: Algorithm to calculate gain in information for the parameters”, Line 20, “Number of nearest neighbours to used for mutual information estimation: k”, P. 22, 7.1.1. “the number of nearest neighbours is k = 10”, P. 24, “A number of Nens = 10000 samples is generated and 25 neighbours are used to approximate the entropies”, D1, P. 729, “Algorithm 3: Generic Particle Filter”, resampling occur within the recursion, so the particles at next steps are the resamples ones).
Regarding Claim 9,
Balakrishnan-Pant-D1 teaches the apparatus of claim 1, wherein the action is selected based on expected values of the at least one differential entropy (Balakrishnan, ¶29, “employ an information gain-based criterion. This criterion approximates the expected change in entropy or information gain between the user parameter distribution before and after receiving feedback, for any pair of items. This reduces the task of choosing a feedback pair to simply finding a pair that maximizes the information gain”, ¶30, “the (differential) entropy h for a k-dimensional MVN distribution with covariance matrix S, is given by …”, ¶31, “ the choice of the next pair of items (l, r) to get feedback for would be ones that maximize IGlr”) estimated for the at least one observation caused by the action (Balakrishnan, ¶29, “expected change in entropy or information gain between the user parameter distribution before and after receiving feedback”, ¶30, “given any one pair of items l and r, its pair parameters θlr and pairwise binary response yl>r, the log-likelihood of the user parameters 30 can be obtained using Equation #5 … Using a local quadratic approximation of the expected likelihood …”, ¶21, “ recommender application 26 sends the one or more pairwise questions 38 to the client device 22, and the client device 22 sends the user's responses …”, ¶¶38-39, “Starting with a prior distribution for the user parameter vector, the IG criterion is used to find a pair of items and a feedback is sought for them. The pairwise response is combined with the prior distribution using Bayes rule, employing the likelihood given in Equation #5. This results in the posterior distribution “, system selects the action using an information gain criterion that approximates the expected change in differential entropy or the user parameter distribution before and after receiving the user’s feedback (¶¶29-31). Because the expectation is taken over the possible outcomes of that action, and those outcomes are explicitly binary, this is selecting an action based on expected entropy over all observations caused by the action). The same motivation to combine for claim 1 equally applies for current claim
Regarding Claim 10,
Claim 10 recites analogous limitations to claim 1, therefore claim 10 is rejected based on the same rationale as claim 1.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences).
Regarding Claim 11,
Claim 11 recites analogous limitations to claim 2, therefore claim 11 is rejected based on the same rationale as claim 2.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences).
Regarding Claim 12,
Claim 12 recites analogous limitations to claim 3, therefore claim 12 is rejected based on the same rationale as claim 3.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences).
Regarding Claim 18,
Claim 18 recites analogous limitations to claim 9, therefore claim 18 is rejected based on the same rationale as claim 9.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31 teaches predicting user preferences).
Regarding Claim 19,
Claim 19 recites analogous limitations to claim 1, therefore claim 19 is rejected based on the same rationale as claim 1.
Balakrishnan et al. teaches A computer program product for implementing a computer system to predict preferences, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform operations comprising (see Col. 15 lines 42-50; ¶31, teaches predicting user preferences).
Regarding Claim 20,
Claim 20 recites analogous limitations to claim 9, therefore claim 20 is rejected based on the same rationale as claim 9.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31 teaches predicting user preferences).
Claims 4 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Balakrishnan et al. (US 2012/0143802 A1, hereinafter Balakrishnan) in view of Pant et al. (“An information-theoretic approach to assess practical identifiability of parametric dynamical systems”, hereinafter Pant) Disclosed in IDS submitted 9/25/2025 and further in view of “A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking” Published 2002 [hereinafter D1] further in view of Ajgl et al. (“Differential entropy estimation by particles”) disclosed in IDS.
Regarding Claim 4,
Balakrishnan-Pant-D1 teaches the apparatus of claim 3, wherein the at least one processor device is further configured to estimate the at least one differential entropy of the at least one posterior distribution by approximating a probability density function of the prior distribution at each sample of the second set of samples (Balakrishnan, Fig. 5 teaches processor and memory, Pant, pg. 67 Section 2:
PNG
media_image2.png
405
592
media_image2.png
Greyscale
PNG
media_image3.png
217
597
media_image3.png
Greyscale
teaches estimating the differential entropy of the posterior distribution by updating the prior probability distribution to the posterior distribution using Bayes’ theorem and using the prior distribution’s probability density function px(x), D1, P. 729, “Algorithm 3: Generic Particle Filter”, resampling occurs within the recursion, so the particles at next steps are the resamples ones).
Balakrishnan–Pant-D1 does not appear to explicitly teach using a volume of a sphere having a radius equal to the distance.
However, Ajgl et al. teaches using a volume of a sphere having a radius equal to the distance (pg. 11993 ¶1: “Another way is to use the volume of nx dimensional sphere with the radius
p
1
i
, i.e. the distance to the nearest neighbour” teaches a sphere with radius equal to the distance).
Balakrishnan–Pant-D1 and Ajgl et al. are analogous art because they are directed to analysis related to distributions.
It would have been obvious for one of ordinary skill in the arts before the effective filing date of the claimed invention to incorporate using a volume of a sphere having a radius equal to the distance as taught by Ajgl et al. to the disclosed invention of Balakrishnan–Pant-D1.
One of ordinary skill in the arts would have been motivated to make this modification in order to provide nonparametric entropy estimator that can be applied in the fusion problem, while conventional running particle filter approaches lack such application (Ajgl et al. pg. 11992 ¶2).
Regarding Claim 13,
Claim 13 recites analogous limitations to claim 4, therefore claim 13 is rejected based on the same rationale as claim 4.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences).
Claims 5 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Balakrishnan et al. (US 2012/0143802 A1, hereinafter Balakrishnan) in view of Pant et al. (“An information-theoretic approach to assess practical identifiability of parametric dynamical systems”, hereinafter Pant) Disclosed in IDS submitted 9/25/2025 and further in view of “A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking” Published 2002 [hereinafter D1] further in view of Gupta et al. (“Parametric Bayesian Estimation of Differential Entropy and Relative Entropy”) disclosed in IDS.
Regarding Claim 5,
Balakrishnan–Pant-D1 teaches the apparatus of claim 1.
Balakrishnan et al. further teaches wherein the at least one processor device is further configured to (Fig. 5 teaches processor and memory). The same motivation to combine for claim 1 equally applies for current claim.
Balakrishnan-Pant-D1 does not appear to explicitly teach estimate the at least one differential entropy of the at least one posterior distribution having Euler's constant as a constant term.
However, Gupta et al. teaches estimate the at least one differential entropy of the at least one posterior distribution having Euler's constant as a constant term (pg. 826 ¶-1: “we estimate the differential entropy as: EN[h(N)], where the expectation is taken with respect to the posterior distribution over N” teaches that differential entropy is a characteristic of posterior distribution; pg. 821 ¶-3: “The special case that best validates the high-rate quantization assumptions is when the number of quantization cells is as large as possible, and they show that this special case produces the nearest-neighbor differential entropy estimator originally proposed by Kozachenko and Leonenko in 1987 [9]:
PNG
media_image10.png
94
1153
media_image10.png
Greyscale
where 𝛌 is the Euler-Mascheroni constant” teaches that an estimator for differential entropy (characteristic of posterior distribution) has a Euler-Mascheroni constant, which corresponds to Euler’s constant).
Balakrishnan–Pant-D1 and Gupta et al. are analogous art because they are directed to analysis related to posterior distributions.
It would have been obvious for one of ordinary skill in the arts before the effective filing date of the claimed invention to incorporate estimate the at least one differential entropy of the at least one posterior distribution having Euler's constant as a constant term as taught by Gupta et al. to the disclosed invention of Balakrishnan–Pant-D1.
One of ordinary skill in the arts would have been motivated to make this modification in order to provide an approach for estimation of differential entropy and relative entropy that has significant performance improvement over other estimates approaches (Gupta et al. pg. 819 ¶-2).
Regarding Claim 14,
Claim 14 recites analogous limitations to claim 5, therefore claim 14 is rejected based on the same rationale as claim 5.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences).
Claims 6 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Balakrishnan et al. (US 2012/0143802 A1, hereinafter Balakrishnan) in view of Pant et al. (“An information-theoretic approach to assess practical identifiability of parametric dynamical systems”, hereinafter Pant) Disclosed in IDS submitted 9/25/2025 and further in view of “A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking” Published 2002 [hereinafter D1] further in view of Suzuki (US 2011/0060708 A1) Disclosed in IDS submitted 9/25/2025.
Regarding Claim 6,
Balakrishnan–Pant-D1 teaches the apparatus of claim 1 , wherein the at least one processor device is further configured to (Balakrishnan Fig. 5 teaches processor and memory), estimate the at least one differential entropy of each posterior distribution of a plurality of posterior distributions (P. 68, “the posterior probability distribution of the parameters evolves as follows by successive observations Zj = zj
PNG
media_image11.png
76
602
media_image11.png
Greyscale
, “For the above probability distributions, the differential entropy evolves as
PNG
media_image12.png
102
655
media_image12.png
Greyscale
P. 72, Algorithm1, lines 23-26, P. 67, “For a continuous random variable X with a probability density function pX(x), the analogue of Shannon entropy, often referred as differential entropy is defined as … Eq. (1)) based on the parameter relating to the density of the prior distribution at each sample of the second set of samples (P. 71,
PNG
media_image13.png
500
816
media_image13.png
Greyscale
) and a likelihood of transition for each sample of the second set of samples from the prior distribution to each posterior distribution of the plurality of posterior distributions, , and the plurality of posterior distributions includes the at least one posterior distribution (P. 72, Algorithm1, lines 23-26, pg. 67 Section 2:
PNG
media_image2.png
405
592
media_image2.png
Greyscale
PNG
media_image3.png
217
597
media_image3.png
Greyscale
teaches estimating the differential entropy of the posterior distribution by updating the prior probability distribution to the posterior distribution using Bayes’ theorem and using the probability density function px(x) and the likelihood of py|x(y|x), which corresponds to estimating the differential entropy of a posterior distribution based on parameter related to density and likelihood of transition, P. 71, Eq. 29-31, D1, P. 7, Algorithm 3).
Balakrishnan–Pant-D1 are analogous art because they are directed to analysis related to posterior distributions.
It would have been obvious for one of ordinary skill in the arts before the effective filing date of the claimed invention to incorporate estimate the at least one differential entropy of each of a plurality of posterior distributions based on the at least one parameter relating to the density at each sample and a likelihood of transition for each sample from the prior distribution to each posterior distribution as taught by Pant to the disclosed invention of Balakrishnan–Pant-D1
One of ordinary skill in the arts would have been motivated to make this modification in order to provide a framework for quantification of information gain measurements that is easily parallelisable which allows for efficient, scalable analysis of large datasets (Pant pg. 66-67 Section 1). This is simply combining prior art elements according to known methods to yield predictable results and applying a known technique to a known device (method, or product) ready for improvement to yield predictable results (MPEP 2143).
Balakrishnan–Pant-D1 does not appear to explicitly teach wherein each likelihood of transition exceeds a threshold likelihood.
However, Suzuki teaches wherein each likelihood of transition exceeds a threshold likelihood (Fig. 10, Fig. 31 S182: “Set a transition probability equal to or greater than a threshold (here, 0.01) to 0.9…” teaches a transition probability exceeding a threshold probability; Fig. 46 teaches the input data to the ACHMM includes samples, ¶¶239-241).
Balakrishnan–Pant-D1 and Suzuki are analogous art because they are directed to analysis related to posterior distributions.
It would have been obvious for one of ordinary skill in the arts before the effective filing date of the claimed invention to incorporate wherein each likelihood of transition exceeds a threshold likelihood as taught by Suzuki to the disclosed invention of Balakrishnan–Pant-D1.
One of ordinary skill in the arts would have been motivated to make this modification in order to provide a system that enables improvement to the posterior probability of the ACHMM (Suzuki pg. 44 [0944]).
Regarding Claim 15,
Claim 15 recites analogous limitations to claim 6, therefore claim 15 is rejected based on the same rationale as claim 6.
Balakrishnan et al. teaches A computer-implemented method for implementing a computer system to predict preferences, comprising (Fig. 5 teaches processor and memory; ¶31, teaches predicting user preferences).
Response to Arguments
Applicant arguments regarding how the amended claims represent an improvement to the technology are persuasive therefore, the examiner respectfully withdraw the 35 USC 101 rejection (Remarks P. 13-16).
Applicant argues that Balakrishan merely determines a likelihood based on a model of a user's pairwise preference between items. However, Balakrishan does not teach or suggest obtaining, from a behavioral model of at least one person, a likelihood of observation for each sample of a plurality of samples, where the behavioral model has an internal state estimated by the prior distribution and the likelihood of observation of each sample of a first set of samples of the plurality of samples is less than a threshold value (Remarks P. 19).
Examiner respectfully disagrees, applicant arguments is related to Balakrishan alone, but the rejection is an obvious rejection. Balakrishan is cited to teach a behavior model of a person and its internal state See at least ¶16 that disclose the latent factor model, user specific parameters whose combination determines the preference a user will have for item, estimated by eq. 2, and refined by each successive response ¶18, eq. 5. D1 is relied on to modify Balakrishan to include drawing a plurality of samples from the prior See at least Alg. 4; and evaluating the likelihood of the observation for each sample, then eliminating those samples whose likelihood falls below a level. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Applicant argues that Balakrishan nowhere teaches or suggests using a behavioral model generated likelihood to eliminate the first set of samples prior to an entropy determination. To the contrary, Balakrishan expressly relies on likelihood terms to construct and approximate posterior distribution (Remarks P. 19).
Examiner respectfully disagrees, applicant arguments are related to Balakrishan alone, but the rejection is an obvious rejection. D1 is relied on to modify Balakrishan to include drawing a plurality of samples from the prior See at least Alg. 4, and evaluating the likelihood of the observation for each sample, then eliminating those samples whose likelihood falls below a level. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Applicant argues that D1 merely describes resampling in a particle-filtering framework when a degeneracy condition or threshold condition is satisfied, where particles having small weights are eliminated, to concentrate on particles with larger weights. However, D1 does not teach or suggest obtaining, from a behavioral model of at least one person, a likelihood of observation for each sample of the plurality of samples, where the likelihood of observation of each sample of a first set of samples of the plurality of samples is less than a threshold value. Further, D1 does not teach or suggest eliminating the first set of samples from the plurality of samples to generate a second set of samples from the plurality of samples (Remarks P. 19-20).
Examiner respectfully disagrees, applicant arguments are related to D1 alone, but the rejection is an obvious rejection. Balakrishan is cited to teach a behavior model of a person and its internal state See at least ¶16 that disclose the latent factor model, user specific parameters whose combination determines the preference a user will have for item, estimated by eq. 2, and refined by each successive response ¶18, eq. 5. D1 is relied on for the sampling operation. As to the likelihood of observation for each sample, D1 disclose sampling importance resampling (SIR) filter with the prior as importance density and resampling at every index, the particle weight reduces Eq. (66), computed for each particle Alg. 4 which is the likelihood of the observation given that sample. D1 further generate a second set as Alg. 2 generates a new set by resampling from the discrete representation of eq. (64). In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Applicant argues that D1 operates in a fundamentally different framework than the claimed invention. In particular, the amended independent claim 1 is directed to a behavioral model having an internal state estimated using a prior distribution, where the behavioral model outputs an observation likelihood for each sample, and where samples having observation likelihood less than a threshold are eliminated prior to determining differential entropy. Thus, the observation likelihood is generated by the behavioral model and is specifically used as a gating criterion to determine which samples are eligible to contribute to the differential entropy determination
Examiner respectfully disagrees, applicant arguments are related to D1 alone, but the rejection is an obvious rejection. Balakrishan is cited to teach a behavior model of a person and its internal state See at least ¶16 that disclose the latent factor model, user specific parameters whose combination determines the preference a user will have for item, estimated by eq. 2, and refined by each successive response ¶18, eq. 5. D1 is relied on to modify Balakrishan to include drawing a plurality of samples from the prior See at least Alg. 4, and evaluating the likelihood of the observation for each sample, then eliminating those samples whose likelihood falls below a level. Pant further disclose the differential entropy estimated from surviving ensemble (6.1, Eq. 29-31. The argument related to the different framework is not persuasive as Balakrishan operates in a Bayesian framework in which a prior is updated to a posterior via Bayer’s rule using a likelihood, and the posterior becomes the prior for the next step ¶38 which is the same framework of D1 Eq. (4). Examiner further notes that the claim recited relationship is that the density and entropy operations are performed on the second set, not any temporal sequencing and the combination satisfies it because Pant’s estimator runs over ensemble D1 resampling that as shown in Alg. 3 places resampling within the recursion, so the ensemble carried forward and over with Pant estimator runs is necessarily the resampled set. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Applicant argues that D1 does not teach or suggest a behavioral model. Rather, the D1 is directed to generic Bayesian state estimation using particle filters in which a likelihood function is derived from a measurement model. The likelihood in D1 is not an output of a behavioral model having an internally estimated state. Instead, the likelihood in D1 is merely one component of a posterior probability distribution approximation. D1 repeatedly teaches that particles are assigned weights based on Bayesian filtering equations and that particles having relatively insignificant weights are subsequently be removed through resampling.
Examiner respectfully disagrees; the rejection does not rely on D1 to teach the argued limitation. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Applicant argues that because D1 neither discloses a behavioral model having an internal state estimated from a prior distribution nor uses a behavioral-model-generated observation likelihood as a criterion for excluding samples from an entropy calculation, D1 provides no teaching, suggestion, or motivation to modify its particle-filter framework to arrive at the claimed invention. Any conclusion that D1's weight-based particle resampling renders obvious the claimed observation-likelihood-based sample elimination would therefore require impermissible hindsight reconstruction using Applicant's disclosure as a roadmap.
Examiner respectfully disagrees; the rejection does not rely on D1 to teach a behavioral model having an internal state. D1 is relied on only for evaluating a likelihood of observation for each sample and eliminating samples based on the evaluation. The motivation to combine Balakrishan and D1 is drawn from the references. Balakrishan states that non-conjugacy leaves no closed form posterior and require approximation ¶40, and D1 discloses that degeneracy devotes computation to particles whose contribution is almost zero P.179. Reliance on references own stated problems and solutions are not hindsight.
Applicant argues that even if Balakrishan were combined with D1, the resulting system would at most suggest using likelihoods to update a posterior distribution and using resampling techniques to mitigate particle degeneracy. Neither reference, alone or in combination, teaches or suggests a behavioral model having an internal state estimated by a prior distribution, the behavioral model outputting an observation likelihood, and the observation likelihood being used to eliminate samples from consideration before determining differential entropy.
Examiner respectfully disagrees; the combined references teach the argued limitations. Balakrishan is cited to teach a behavior model of a person and its internal state See at least ¶16 that disclose the latent factor model, user specific parameters whose combination determines the preference a user will have for item, estimated by eq. 2, and refined by each successive response ¶18, eq. 5. D1 is relied on to modify Balakrishan to include drawing a plurality of samples from the prior See at least Alg. 4, and evaluating the likelihood of the observation for each sample, then eliminating those samples whose likelihood falls below a level. Pant further disclose the differential entropy estimated from surviving ensemble (6.1, Eq. 29-31. The argument related to the different framework is not persuasive as Balakrishan operates in a Bayesian framework in which a prior is updated to a posterior via Bayer’s rule using a likelihood, and the posterior becomes the prior for the next step ¶38 which is the same framework of D1 Eq. (4). Examiner further notes that the claim recited relationship is that the density and entropy operations are performed on the second set, not any temporal sequencing and the combination satisfies it because Pant’s estimator runs over ensemble D1 resampling that as shown in Alg. 3 places resampling within the recursion, so the ensemble carried forward and over with Pant estimator runs is necessarily the resampled set.
Applicant argues that because Balakrishan and D1 employ likelihoods for different purposes and in different estimation frameworks, there is no articulated reasoning or motivation that would have led a skilled artisan to modify either reference to perform the claimed likelihood-based sample elimination prior to differential entropy determination
Examiner respectfully disagrees, the motivation to combine Balakrishan and D1 is drawn from the references. Balakrishan states that non-conjugacy leaves no closed form posterior and require approximation ¶40, and D1 discloses that degeneracy devotes computation to particles whose contribution is almost zero P.179. Further, Pant discloses that analytical tractability of the probability distributions and their integration to calculate entropy is a concern, especially when the forward and observation operators are non-linear and dimensionality of the problem is high. Nonetheless, one may be able to simulate a large, yet finite, number of particles (samples from the prior distribution) forward in time to obtain samples from the probability distribution of Z1:j. For this reason, methods to estimate H(X) only from samples of X are sought (Pant et al. pg. 66-67 Section 1, pg. 70-71 Section 6). This is simply combining prior art elements according to known methods to yield predictable results and applying a known technique to a known device (method, or product) ready for improvement to yield predictable results (MPEP 2143).
Applicant argues that that the combination of references Balakrishnan, Pant, and D1 does not teach or suggest at least, for example, the features of "obtain, from a behavioral model of at least one person, a likelihood of observation for each sample of the plurality of samples the behavioral model has an internal state estimated by the prior distribution eliminate a first set of samples from the plurality of samples to generate a second set of samples from the plurality of samples the likelihood of observation of each sample of the first set of samples is less than a threshold value"
Examiner respectfully disagrees, applicant arguments are related to Balakrishan alone, but the rejection is an obvious rejection. Balakrishan is cited to teach a behavior model of a person and its internal state See at least ¶16 that disclose the latent factor model, user specific parameters whose combination determines the preference a user will have for item, estimated by eq. 2, and refined by each successive response ¶18, ¶25 eq. 5. D1 is relied on to modify Balakrishan to include drawing a plurality of samples from the prior See at least Alg. 4, and evaluating the likelihood of the observation for each sample, then eliminating those samples whose likelihood falls below a level.
Conclusion
The prior art made of record and not relied upon is considered pertinent to the applicant' s disclosure.
Osogami et al. “A hierarchical Bayesian choice model with visibility” that teach a hierarchical Bayesian choice model to estimate customer preference and visibility of items.
Examiner has pointed out particular references contained in the prior arts of record in the body of this action for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and Figures may apply as well. It is respectfully requested from the applicant, in preparing the response, to consider fully the entire references as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior arts or disclosed by the examiner. It is noted that any citation to specific pages, columns, figures, or lines in the prior art references any interpretation of the references should not be considered to be limiting in any way. A reference is relevant for all it contains and may be relied upon for all that it would have reasonably suggested to one having ordinary skill in the art. In re Heck, 699 F.2d 1331-33, 216 USPQ 1038-39 (Fed. Cir. 1983) (quoting In re Lemelson, 397 F.2d 1006, 1009, 158 USPQ 275, 277 (CCPA 1968)).
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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOHAMED ABOU EL SEOUD whose telephone number is (303)297-4285. The examiner can normally be reached Monday-Thursday 9:00am-6:00pm MT.
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, Michelle Bechtold can be reached at (571) 431-0762. 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.
/MOHAMED ABOU EL SEOUD/Primary Examiner, Art Unit 2148
Prosecution Insights