DETAILED ACTION
Claims 1, 4-9, 12-17, and 20 are presented for examination.
This Office Action is in response to submission of documents on January 16, 2026.
Rejection of claims 1-2, 4-8, 10-18, and 20 under 35 U.S.C. 101 for being directed to unpatentable subject matter is withdrawn.
Rejection of claims 1-2, 4-8, 10-18, and 20 under 35 U.S.C. 103 as being obvious over Melkumyan in view of Noack is withdrawn.
New rejection of claims 1, 4-9, 12-17, and 20 under 35 U.S.C. 103 as being obvious over Melkumyan in view of Han and Noack.
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 .
Response to Arguments
Regarding rejection of the claims under 35 U.S.C. 101, Examiner agrees that the claims, as amended, overcome the rejection under 35 U.S.C. 101. Accordingly, the rejection is withdrawn.
Regarding the rejection of the claims under 35 U.S.C. 112(a), Examiner agrees that the claims, as amended, overcome the rejection. Accordingly, the rejection under 35 U.S.C. 112(a) is withdrawn.
Regarding rejection of the claims under 35 U.S.C. 103, Examiner agrees that Melkumyan fails to teach or suggest “determine a covariance between a selected data value and data values in the dataset using the covariance function, the selected data value being a representative data point representing a cluster of data points,” as currently presented. Particularly, Melkumyan does not appear to explicitly disclose “the selected data value being a representative data point representing a cluster of data points.” However, Han, et al., (“Data Pre-processing”) discloses representing a cluster of points with a centroid, which is representative of the cluster. Accordingly, rejection of the claims over Melkumyan in view of Noack is withdrawn and a new rejection of the claims is asserted under 35 U.S.C. 103 as being obvious over Melkumyan in view of Han and Noack.
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 (i.e., changing from AIA to pre-AIA ) 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-2,4-10,12-18,20 are rejected under 35 U.S.C. 103 as being obvious over Melkumyan, et al., (U.S. Pat. No. 8,849,622, hereinafter “Melkumyan”) in view of Han, et al., (“Data Pre-processing,” hereinafter “Han”) and Noack, et al., (“Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels,” hereinafter “Noack”).
Claim 1
Melkumyan discloses:
A system comprising: a processor; and
The measurement sensor data generated by the sensors 230 is provided to a training processor 240 coupled to data storage 250. Melkumyan at col. 5, lines 17-18.
The computing system 100 comprises suitable components necessary to receive, store and execute appropriate computer instructions. The components may include a processing unit 102, read only memory (ROM) 104, random access memory (RAM) 106… Melkumyan at col. 4, lines 7-9.
a memory storing machine-readable instructions that, when executed by the processor, cause the processor to:
The training processor 240 is adapted to organise the sensor data and determine a non-parametric, probabilistic, multi-scale representation of the data for use in terrain modelling, which is stored in the data storage 250. Melkumyan at col. 5, lines 19-22.
The computing system 100 comprises suitable components necessary to receive, store and execute appropriate computer instructions. The components may include a processing unit 102, read only memory (ROM) 104, random access memory (RAM) 106… Melkumyan at col. 4, lines 7-9.
train a model on a dataset to learn a covariance function;
Training the GP for a given dataset is tantamount to optimising the hyperparameters of the underlying covariance function. This training can be done using machine learning. It can also be done manually, for example by estimating the values and performing an iterative fitting process. Melkumyan at col. 8, lines 56-61.
“GP” is “Gaussian Process.”
determine a covariance between a selected data value and data values in the dataset using the covariance function,
For problems with thousands of observations, exact inference in normal GPs is intractable and approximation algorithms are required. Most of the approximation algorithms employ a subset of points to approximate the posterior distribution of a new point given the training data and hyperparameters. Melkumyan at col. 10, lines 28-33.
The “posterior distribution of a new point” is analogous to a covariance and “hyperparameters” are analogous to the “covariance function.”
The “data values in the dataset” are analogous to “a subset of points.”
select a subset of the dataset, the subset including the data values that have a covariance value that meets or exceeds a predetermined threshold value; and
These approximations rely on heuristics to select the subset of points, or use pseudo targets obtained during the optimization of the log-marginal likelihood of the model. Melkumyan at col. 10, lines 33-36.
A “pseudo target” is analogous to a “predetermined threshold value.”
Melkumyan does not appear to disclose:
the selected data value being a representative data point representing a cluster of data points;
predict at least one characteristic of one or more potential experiments based on the subset, the at least one characteristic being one of a type, an outcome, or an uncertainty level of the one or more potential experiments, the subset comprising fewer data values than the dataset and reducing computational demand associated with prediction of the one or more potential experiments.
Han, which is analogous art, discloses:
the selected data value being a representative data point representing a cluster of data points;
Centroid distance is an alternative measure of cluster quality and is defined as the average distance of each cluster object from the cluster centroid (denoting the “average object,” or average point in space for the cluster). Han at pg. 108.
A centroid of a cluster is representative of a cluster of points because it is the “average point in space for the cluster.”
Han is analogous art to the claimed invention because both include techniques for analysis of a cluster of points. It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to select a centroid to represent a cluster of points, as disclosed in Han, when determining a covariance, as disclosed in Melkumyan. Motivation to combine includes ensuring that the characteristics of the one or more potential experiments is representative of the cluster of points, thereby improving the accuracy of the results.
Noack, which is analogous art, discloses:
predict at least one characteristic of one or more potential experiments based on the subset, the at least one characteristic being one of a type, an outcome, or an uncertainty level of the one or more potential experiments, the subset comprising fewer data values than the dataset and reducing computational demand associated with prediction of the one or more potential experiments.
Gaussian process regression (GPR) techniques have emerged as the method of choice for steering many classes of experiments. We have recently demonstrated the positive impact of GPR-driven decision-making algorithms on autonomously-steered experiments at a synchrotron beamline. Noack at Abstract.
The success of GPR in steering experiments is due to its non-parametric nature; simply speaking, the more data that is gathered the more complicated the model function can become. The number of parameters of the function, and therefore its complexity, does not have to be defined a priori. This is in contrast to neural networks, which need a specification of an architecture (number of layers, layer width, activation function) beforehand. GPR also naturally includes uncertainty quantification, which is an absolute necessity in experimental sciences. Noack at pg. 3.
Noack is analogous art to the claimed invention because both are directed to autonomous design of experiments using a model, particularly a Gaussian process model. It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the application, to combine the training of a Gaussian process model, as disclosed in Melkumyan, with the design of experiments using a Gaussian process, as disclosed in Noack, to result in a system that trains a model using gathered data and predicts experiments using the trained model. Motivation to combine includes reducing the complexity of execution of the model by limiting the data required to predict experiment design, thereby reducing computation time and resources.
Claim 4
Melkumyan discloses:
train the model on the subset.
Training the GP for a given dataset is tantamount to optimising the hyperparameters of the underlying covariance function. This training can be done using machine learning. It can also be done manually, for example by estimating the values and performing an iterative fitting process. Melkumyan at col. 8, lines 56-61.
“GP” is “Gaussian Process.”
The original dataset includes the subset and therefore the step of training the model on the dataset includes training the model on the subset.
Claim 5
Melkumyan does not appear to disclose:
refit the covariance function with the subset.
Noack discloses:
refit the covariance function with the subset.
The variance of real experimental measurements vary greatly across the parameter space, and this has to be reflected in the steering process as well as in the final model creation. For instance, in x-ray scattering experiments, the variance of a raw measurement depends strongly on the exposure time; computed quantities can have wildly different variances depending on the raw data in that part of the space (e.g. fit quality will not be uniform), and material heterogeneity will depend strongly on location within the parameter space. These inhomogeneities in the measurement noise need to be actively included in the final model to avoid interpolation mistakes and consequently erroneous models. Noack at pg. 2, paragraph 5.
“Actively including” inhomogeneities into the model is analogous to “refitting” the covariance function.
Claim 6
Melkumyan discloses:
train a second model with the subset.
The output 530 of the Gaussian process evaluation 520 is a digital elevation map/grid at the chosen resolution and region of interest together with an appropriate measure of uncertainty for every point in the map. The digital elevation map may be used as is or may be rapidly processed into a digital surface/terrain model and used thereafter for robotic vehicle navigation and the like in known fashion. Melkumyan at col. 6, line63-col. 7, line 2.
The “digital surface/terrain model” is a second model.
Claim 7
Melkumyan discloses:
select the dataset from a larger dataset in a random manner.
The inference set contains the points used to perform inference on the testing points. For each case the experiment is repeated 1500 times with randomly selected inference and testing sets. Melkumyan at col. 14, lines 37-40.
Claim 8
Melkumyan discloses:
wherein the model is a Gaussian process model.
In another embodiment the kernel machine uses a Gaussian learning process. Melkumyan at col. 2, lines 61-62.
Claims 9 and 12-16
Claims 9 and 12-16 recite a method that is substantially the same as the method performed by the system recited in claims 1 and 4-8. Accordingly, for at least the same reasons and based on the same prior art as claims 1-2 and 4-8, claims 9 and 12-16 are rejected under 35 U.S.C. 103 as being obvious over Melkumyan in view of Han and Noack.
Claims 17 and 20
Claim 17 recites:
A non-transitory computer-readable medium including instructions that when executed by a processor
The computing system may include storage devices such as a disk drive 108 which may encompass solid state drives, hard disk drives, optical drives or magnetic tape drives. The computing system 100 may use a single disk drive or multiple disk drives. A suitable operating system 112 resides on the disk drive or in the ROM of the computing system 100 and cooperates with the hardware to provide an environment in which software applications can be executed. Melkumyan at col. 4, lines 27-34.
The claim further recites a method stored on the medium that is substantially the same as the method recited in claim 1. Claim 20 disclose substantially the same limitations as claim 8. Accordingly, for at least the same reasons and based on the same prior art as claims 1 and 8, claims 17 and 20 are rejected under 35 U.S.C. 103 as being obvious over Melkumyan in view of Hand and Noack.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Srinivasan, et al., “Efficient subset selection via the kernelized Renyi distance.”
Duplyakin, et al., “Active Learning in Performance Analysis.”
Yeh, et al., “An Empirical Study of the Sample Size Variability of Optimal Active Learning Using Gaussian Process Regression.”
Kloppenburg, U.S. Pat. No. 10,402,739.
Abdolshah, et al., WIPO App. No. 2022/051794.
Middlebrooks, et al., U.S. Pat. Pub. No. 2021/0286270
Qian, et al., “Gaussian Process Models for Computer Experiments With Qualitative and Quantitative Factors.”
Communication
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSEPH MORRIS whose telephone number is (703)756-5735. The examiner can normally be reached M-F 8:30-5:00.
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, Ryan Pitaro can be reached at (571) 272-4071. 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.
JOSEPH MORRIS
Examiner
Art Unit 2188
/JOSEPH P MORRIS/Examiner, Art Unit 2188
/RYAN F PITARO/Supervisory Patent Examiner, Art Unit 2188
Prosecution Insights