Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Acknowledgment is made of applicant's claim for foreign priority based on an application filed in China on 11/26/2021. It is noted, however, that applicant has not filed a certified copy of the CN 20211416769.8 application as required by 37 CFR 1.55. Please note the failure status report in the application’s file wrapper issued 07/14/2023.
Specification
The abstract of the disclosure is objected to because it contains more than 150 words. A corrected abstract of the disclosure is required and must be presented on a separate sheet, apart from any other text. See MPEP § 608.01(b).
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
Claim Objections
Claims 2, 7, 10, and 13-16 are objected to because of the following informalities: In claim 2, line 4, the limitation “corresponding respective” should recite “corresponding to respective”.
In claim 7, line 1, the limitation “the acquiring at least indicator data in comprises” is not grammatically correct.
In claim 7, line 6, “target scenarios is based on” should recite “target scenarios based on”.
In claim 10, Examiner suggests modifying the limitation in lines 8-9 to read “second determining code configured to, subsequent to obtaining the respective detection results, cause…”
In claim 13, Examiner suggests modifying the limitations in lines 2-3 in a similar way as the suggestion for claim 10 above. In claim 13, line 7, “selecting” should recite “select”.
In claim 14, line 2, “to acquiring” should recite “to acquire”.
In claim 14, Examiner suggests modifying the limitations in line 4 in a similar way as the suggestion for claim 10 above.
In claim 15, line 1, the limitation “computer-readable storing instructions” should recite “computer-readable medium storing instructions” based on the preamble of claim 16.
Claim 16 recites the same minor informality as claim 2. Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
In claim 1, the limitations “at least one indicator data” in line 3 renders the claim indefinite for the following reasons. Lines 7-10 disclose reference indicator data and non-reference indicator data. It is unclear if “at least one indicator data” means one data point having both reference and non-reference indicator data. If so, it is unclear how detection results from a single data point could have different measures of uncertainty. It is unclear if each indicator data point is either a reference data point or a non-reference data point. The limitation in lines 7-10 further renders the claim indefinite because it is unclear if two uncertainties of different detection results have been determined in order to identify a higher uncertainty.
Claim 1 recites the limitation "the detection results" in line 6. There is insufficient antecedent basis for this limitation in the claim. It is noted that claim 1 recites “a detection result” in line 4 but not a plurality of detection results.
Examiner treats each indicator data point in line 3 as being either a reference indicator data point or a non-reference indicator data point. Examiner treats the limitation “respective uncertainty of a detection result” in line 4 as “respective uncertainties of detection results”. Examiner treats each the method as determining a different uncertainty for each detection result.
Claims 2-8 are rejected for failing to cure the deficiencies of claim 1. These claims should be carefully reviewed and amended to align with amendments made to claim 1.
Claim 8 recites the limitation "the plurality of target scenarios" in line 1. There is insufficient antecedent basis for this limitation in the claim. It is unclear if claim 8 is supposed to depend on claim 7, which would provide sufficient antecedent basis for this limitation. Examiner treats the preamble of claim 8 as “The method according to claim 7”.
Claim 9 is an apparatus which recites the same indefinite limitations as the method of claim 1 and is therefore rejected for at least the same reasons.
Claims 10-14 are rejected for failing to cure the deficiencies of claim 9.
Claim 12 recites the limitation "the first selecting data" in line 1. There is insufficient antecedent basis for this limitation in the claim. It is unclear if this limitation is supposed to recite “the first selecting code”, which would have sufficient antecedent basis in claim 9 on page 4, line 1. Examiner treats “the first selecting data” as “the first selecting code”.
In claim 14, the limitation “second acquiring code” in line 2 renders the claim indefinite. Parent claim 9 recites “second acquiring code” on page 4, line 6. It is unclear whether “second acquiring code” in claims 9 and 14 are the same code or different codes. Examiner treats them as the same code, and treats this limitation in claim 14 as “the second acquiring code”.
Claim 15 is a product which recites the same indefinite limitations as the method of claim 1 and is therefore rejected for at least the same reasons.
Claims 16-20 are rejected for failing to cure the deficiencies of claim 15.
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 is directed to an abstract idea without significantly more.
Step 1: Claims 1-8 each recites a method, claims 9-14 each recites an apparatus comprising a processor, and claims 15-20 each recites a non-transitory computer-readable medium (a product). A method, an apparatus, and a product each falls under one of four statutory categories of patent eligible subject matter.
Claim 1
Step 2A Prong 1: Determining,
Selecting reference indicator data from the at least one indicator data based on the respective uncertainty of the detection result, wherein an uncertainty of the detection result corresponding to the reference indicator data is higher than the respective uncertainty of the detection result corresponding to non-reference indicator data among the at least one indicator data is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper. The claim recites an abstract idea.
Step 2A Prong 2: Optimizing training a deep neural network model amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
The method being executed by at least one processor amounts to generic computer components for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Acquiring at least one indicator data amounts to an insignificant pre-solution activity under MPEP 2106.05(g).
Applying the deep neural network model amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Acquiring label detection results corresponding to the reference indicator data amounts to an insignificant post-solution activity under MPEP 2106.05(g).
Obtaining a trained target indicator detection model based on training the deep neural network model based on the reference indicator data and the label detection results corresponding to the reference indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
The additional elements as disclosed above, alone or in combination, do not integrate the abstract ideas into a practical application as they are mere generic computer functions as disclosed in combination with insignificant extra solution activities that are implemented to perform the abstract ideas disclosed above. The claim is directed to an abstract idea.
Step 2B: Optimizing training a deep neural network model amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
The method being executed by at least one processor amounts to generic computer components for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Acquiring at least one indicator data amounts to retrieving data from memory, which courts have recognized as well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
Applying the deep neural network model amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Acquiring label detection results corresponding to the reference indicator data amounts to retrieving data from memory, which courts have recognized as well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
Obtaining a trained target indicator detection model based on training the deep neural network model based on the reference indicator data and the label detection results corresponding to the reference indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
The additional elements as disclosed above, in combination with the abstract ideas, are not sufficient to amount to significantly more than the abstract ideas as they are generic computer functions as disclosed in combination with well-understood, routine and conventional activities that are implemented to perform the abstract ideas disclosed above. The claim is not patent eligible.
Claim 2 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. The determining of the respective uncertainty of a detection result corresponding respective indicator data comprises: … subsequent to obtaining the respective detection results, determining the respective uncertainty of the detection results corresponding to the respective indicator data based on the respective detection results is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Step 2A Prong 2: The deep neural network model is a random dropout neural network model, wherein the random dropout neural network model, during operation, randomly drops inside neuron connections based on a preset dropout rate amounts to an insignificant extra-solution activity under MPEP 2106.05(g).
Obtaining, by the random dropout neural network model, respective detection results based on multi-time forward propagation on the respective indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Step 2B: The deep neural network model is a random dropout neural network model, wherein the random dropout neural network model, during operation, randomly drops inside neuron connections based on a preset dropout rate amounts to well-understood, routine, conventional activity under MPEP 2106.05(d)(I). Gal et al. (“Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning”) provides Berkheimer evidence. Page 2, lines 2-3 discloses “Dropout is used in many models in deep learning as a way to avoid over-fitting”, and page 3, col. 1, lines 6-13 and lines 1-6 below equation 1 discloses the details of dropout as claimed . A preset dropout rate is pi.
Obtaining, by the random dropout neural network model, respective detection results based on multi-time forward propagation on the respective indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 3 incorporates the rejection of claim 2.
Step 2A Prong 1: The abstract ideas of claim 2 are incorporated. The determining of the respective uncertainty of the detection results subsequent to obtaining the respective detection results comprises: determining at least one of a detection result distribution variance and a detection result distribution standard deviation of the respective detection results corresponding to the multi-time forward propagation is a mathematical calculation. Specification paragraph [0063] and formula (7) discloses calculating a prediction variance.
Determining the respective uncertainty of the detection results corresponding to the respective indicator data based on the at least one of the detection result distribution variance and the detection result distribution standard deviation is a mathematical calculation. Specification paragraph [0063] and formula (7) discloses calculating a prediction variance, and a standard deviation is a square root of a variance.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas. The claim is not patent eligible.
Claim 4 incorporates the rejection of claim 3.
Step 2A Prong 1: The abstract ideas of claim 3 are incorporated. Determining a detection result mean based on the respective detection results corresponding to the multi-time forward propagation is a mathematical calculation. Specification paragraph [0063] and formula (5) discloses calculating a predicted mean.
Determining the respective detection results corresponding to the respective indicator data based on the detection result mean is a judgement mental process based on a mathematical calculation which can reasonably be performed in the human mind with the aid of pencil and paper.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas. The claim is not patent eligible.
Claim 5 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. The selecting the reference indicator data comprises at least one of: based on a first uncertainty of the detection result corresponding to a first indicator data exceeding a preset threshold, selecting the first indicator data as the reference indicator data is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Ranking the respective uncertainty of the detection result corresponding to the respective indicator data from high to low is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Selecting a preset quantity of indicator data ranking higher than a ranking threshold as the reference indicator data is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Step 2A Prong 2 and Step 2B: The claim does not recite any additional elements which, alone or in combination, would integrate the abstract ideas into a practical application or which, in combination with the abstract ideas, would be sufficient to amount to significantly more than the abstract ideas. The claim is not patent eligible.
Claim 6 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. For respective updated indicator data, determining the respective uncertainty of a detection result corresponding to the respective updated indicator data based on the target indicator detection model is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Selecting updated reference indicator data from among the at least one updated indicator data based on the respective uncertainty of the detection result corresponding to the respective updated indicator data is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Step 2A Prong 2: Acquiring at least one updated indicator data amounts to an insignificant post-solution activity under MPEP 2106.05(g).
Acquiring label detection results corresponding to the updated reference indicator data amounts to an insignificant post-solution activity under MPEP 2106.05(g).
Obtaining an updated target indicator detection model by training the target indicator detection model based on the updated reference indicator data and the label detection results corresponding to the updated reference indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Step 2B: Acquiring at least one updated indicator data amounts to retrieving data from memory, which courts have recognized as well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
Acquiring label detection results corresponding to the updated reference indicator data amounts to retrieving data from memory, which courts have recognized as well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
Obtaining an updated target indicator detection model by training the target indicator detection model based on the updated reference indicator data and the label detection results corresponding to the updated reference indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 7 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated. Determining a plurality of target scenarios is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
The selecting the reference indicator data comprises: selecting the reference indicator data from the at least one indicator data for the plurality of target scenarios is based on the respective uncertainty of the detection result corresponding to the respective indicator data for the plurality of target scenarios is a judgement mental process which can reasonably be performed in the human mind with the aid of pencil and paper.
Step 2A Prong 2: Acquiring the at least one indicator data for the plurality of target scenarios amounts to an insignificant post-solution activity under MPEP 2106.05(g).
The obtaining the target indicator detection model comprises: obtaining the target indicator detection model by training the deep neural network model based on the reference indicator data and label detection results corresponding to the reference indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f).
Step 2B: Acquiring the at least one indicator data for the plurality of target scenarios amounts to retrieving data from memory, which courts have recognized as well-understood, routine, conventional activity under MPEP 2106.05(d)(II).
The obtaining the target indicator detection model comprises: obtaining the target indicator detection model by training the deep neural network model based on the reference indicator data and label detection results corresponding to the reference indicator data amounts to mere instructions to apply the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 8 incorporates the rejection of claim 1.
Step 2A Prong 1: The abstract ideas of claim 1 are incorporated.
Step 2A Prong 2 and Step 2B: The plurality of the target scenarios comprises one or more of: a microservice monitoring scenario, a physical entity monitoring scenario, a logical entity monitoring scenario, a network topology monitoring scenario, or a log data monitoring scenario amounts to a mere field of use and technological environment under MPEP 2106.05(h). The claim is not patent eligible.
Claim 9 recites an apparatus which implements the same features as the method of claim 1 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitations of an apparatus for optimizing training a deep neural network model, the apparatus comprising: at least one first memory configured to store a first program code, and at least one first processor configured to read the first program code and operate as instructed by the first program code amount to generic computer components for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
The first program code comprising: first acquiring code, first determining code, first selecting code, second acquiring code, and first obtaining code amount to mere instructions for applying the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 10 recites an apparatus which implements the same features as the method of claim 2 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitations of the first determining code comprises: second obtaining code and second determining code amount to mere instructions for applying the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 11 recites an apparatus which implements the same features as the method of claim 3 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitations of the second determining code comprises: third determining code and fourth determining code amount to mere instructions for applying the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 12 recites an apparatus which implements the same features as the method of claim 4 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitations of fifth determining code and sixth determining code amount to mere instructions for applying the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 13 recites an apparatus which implements the same features as the method of claim 5 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitation of the first selecting code comprises at least one of: second selecting code or first ranking code amounts to mere instructions for applying the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 14 recites an apparatus which implements the same features as the method of claim 6 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitations of second acquiring code, seventh determining code, third selecting code, third acquiring code, and third obtaining code amount to mere instructions for applying the abstract ideas on a generic computer under MPEP 2106.05(f). The claim is not patent eligible.
Claim 15 recites a product which implements the same features as the method of claim 1 and is therefore rejected for at least the same reasons.
In Step 2A Prong 2 and Step 2B, the limitations of a non-transitory computer-readable storing instructions that cause at least one processor to execute operations amount to generic computer components for applying the abstract ideas on a generic computer under MPEP 2106.05(f).
Claims 16-20 each recites an product which implements the same features as the method of claims 2-6, respectively, and are therefore rejected for at least the same reasons.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1, 5-9, 13-15, and 19-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Susaiyah et al. (US 20250238720 A1).
Regarding claim 1, Susaiyah teaches: A method for optimizing training a deep neural network model, the method being executed by at least one processor, the method comprising: ([0018], lines 1-4 and [0025], lines 22-24 teaches a processor and a neural network model. Optimizing training is taught below.)
acquiring at least one indicator data; ([0032], lines 1-5, [0035], and [0044] disclose acquiring a pool of unlabeled data samples to be input to the first model. Each data sample is an indicator data point (an indicator datum) which indicates a state of a medical condition.)
determining, by the deep neural network model, respective uncertainty of a detection result corresponding to respective indicator data among the at least one indicator data, wherein the respective uncertainty indicates a respective degree of reliability of the detection results; ([0076]-[0080] discloses classifying two instances (unlabeled input data) and determining uncertainty based on entropy. Case II is more uncertain (less reliable) than case I because the class probabilities for case II are more similar than for case I. The limitation “a detection result” is a classification having the highest posterior probability for each instance.)
selecting reference indicator data from the at least one indicator data based on the respective uncertainty of the detection result, wherein an uncertainty of the detection result corresponding to the reference indicator data is higher than the respective uncertainty of the detection result corresponding to non-reference indicator data among the at least one indicator data; ([0077]-[0080] discloses case II has a higher entropy (uncertainty) than case I, and the corresponding instance for case II should be sent for annotation. “Reference indicator data” is the instance for case II.)
acquiring label detection results corresponding to the reference indicator data; and ([0076], lines 5-10)
obtaining a trained target indicator detection model based on training the deep neural network model based on the reference indicator data and the label detection results corresponding to the reference indicator data. ([0076], lines 5-10)
Regarding claim 5, Susaiyah teaches: The method according to claim 1, wherein the selecting the reference indicator data comprises at least one of: based on a first uncertainty of the detection result corresponding to a first indicator data exceeding a preset threshold, selecting the first indicator data as the reference indicator data; or ([0076] discloses comparing confidence of a model output to a threshold confidence level, and the confidence may be described using entropy of the posterior probability. An entropy that falls below the threshold confidence level corresponds to “exceeding a preset threshold” when the objective is to find low-confidence results.)
ranking the respective uncertainty of the detection result corresponding to the respective indicator data from high to low, and selecting a preset quantity of indicator data ranking higher than a ranking threshold as the reference indicator data.
Regarding claim 6, Susaiyah teaches: The method according to claim 1, the method further comprising: acquiring at least one updated indicator data; (Susaiyah’s system triggers an upgrade process when a performance measure for the first model is below a threshold. [0032], lines 1-5, [0035], and [0044] disclose acquiring unlabeled input data for the first model. Here “updated indicator data” means a second set of unlabeled input data.)
for respective updated indicator data, determining the respective uncertainty of a detection result corresponding to the respective updated indicator data based on the target indicator detection model; ([0076]-[0080] discloses classifying two instances (unlabeled input data) and determining uncertainty based on entropy. Based on [0044], the first model has already undergone a round of active training.)
selecting updated reference indicator data from among the at least one updated indicator data based on the respective uncertainty of the detection result corresponding to the respective updated indicator data; ([0077]-[0080] discloses case II has a higher entropy (uncertainty) than case I, and the corresponding instance for case II should be sent for annotation. “Updated reference indicator data” is the instance for case II.)
acquiring label detection results corresponding to the updated reference indicator data; and ([0076], lines 5-10)
obtaining an updated target indicator detection model by training the target indicator detection model based on the updated reference indicator data and the label detection results corresponding to the updated reference indicator data. (Based on [0044] and [0076], lines 5-10, the first model is retrained whenever the performance measure for the first model is below a threshold performance level.)
Regarding claim 7, Susaiyah teaches: The method according to claim 1, wherein the acquiring at least indicator data in comprises: determining a plurality of target scenarios; and ([0032], [0036]-[0039] discloses determining a plurality of target scenarios including medical images and vital sign events.)
acquiring the at least one indicator data for the plurality of target scenarios; ([0038])
the selecting the reference indicator data comprises: selecting the reference indicator data from the at least one indicator data for the plurality of target scenarios is based on the respective uncertainty of the detection result corresponding to the respective indicator data for the plurality of target scenarios; and ([0077]-[0080] discloses case II has a higher entropy (uncertainty) than case I, and the corresponding instance for case II should be sent for annotation. The instances can include medical images and vital sign events.)
the obtaining the target indicator detection model comprises: obtaining the target indicator detection model by training the deep neural network model based on the reference indicator data and label detection results corresponding to the reference indicator data. ([0076], lines 5-10)
Regarding claim 8, Susaiyah teaches: The method according to claim 1, (Examiner treats claim 8 as depending on claim 7)
wherein the plurality of the target scenarios comprises one or more of: a microservice monitoring scenario, a physical entity monitoring scenario, a logical entity monitoring scenario, a network topology monitoring scenario, or a log data monitoring scenario. ([0037]-[0038] discloses the first model may detect arrhythmia events based on electrocardiogram (ECG) data. Since an ECG is log data of a heart, detecting arrhythmia events is a log data monitoring scenario. Since a heart is a physical entity, detecting arrhythmia events is a physical entity monitoring scenario.)
Claim 9 recites an apparatus which implements the same features as the method of claim 1 and is therefore rejected for at least the same reasons.
Susaiyah teaches: An apparatus for optimizing training a deep neural network model, the apparatus comprising: at least one first memory configured to store a first program code, and at least one first processor configured to read the first program code and operate as instructed by the first program code, ([0018], lines 1-7)
the first program code comprising: first acquiring code, first determining code, first selecting code, second acquiring code, first obtaining code ([0018], lines 1-7 discloses a set of instructions, which corresponds to all the code recited in claim 9.)
Claim 13 recites an apparatus which implements the same features as the method of claim 5 and is therefore rejected for at least the same reasons.
Susaiyah teaches: the first selecting code comprises at least one of: second selecting code or first ranking code ([0018], lines 1-7 discloses a set of instructions, which corresponds to all the code recited in claim 13.)
Claim 14 recites an apparatus which implements the same features as the method of claim 6 and is therefore rejected for at least the same reasons.
Susaiyah teaches: the program code further comprising: second acquiring code, seventh determining code, third selecting code, third acquiring code, and third obtaining code ([0018], lines 1-7 discloses a set of instructions, which corresponds to all the code recited in claim 14.)
Claim 15 recites a product which implements the same features as the method of claim 1 and is therefore rejected for at least the same reasons.
Susaiyah teaches: A non-transitory computer-readable storing instructions that cause at least one processor to: [execute operations] ([0018], lines 1-7)
Claims 19-20 each recites a product which implements the same features as the method of claims 5-6, respectively, and are therefore rejected for at least the same reasons.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 2-4, 10-12, and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Susaiyah et al. (US 20250238720 A1) in view of Gal et al. (“Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning”).
Regarding claim 2, Susaiyah teaches: The method according to claim 1,
…
the determining of the respective uncertainty of a detection result corresponding respective indicator data comprises: obtaining, by the
subsequent to obtaining the respective detection results, determining the respective uncertainty of the detection results corresponding to the respective indicator data based on the respective detection results. (In [0078]-[0080], the entropy for each instance is calculated based on the posterior probability measures for the two classes A and B.)
However, Susaiyah does not explicitly teach: wherein the deep neural network model is a random dropout neural network model, wherein the random dropout neural network model, during operation, randomly drops inside neuron connections based on a preset dropout rate; …
obtaining, by the random dropout neural network model, respective detection results based on multi-time forward propagation on the respective indicator data;
But Gal teaches: wherein the deep neural network model is a random dropout neural network model, wherein the random dropout neural network model, during operation, randomly drops inside neuron connections based on a preset dropout rate; and (Page 3, col. 1, lines 13-20 and lines 1-6 below equation 1. A preset dropout rate is pi)
the determining of the respective uncertainty of a detection result corresponding respective indicator data comprises: obtaining, by the random dropout neural network model, respective detection results based on multi-time forward propagation on the respective indicator data; and (Page 3, col. 1, lines 17-18 and page 4, col. 1, start of § 4 to line 4 below equation 6 discloses obtaining T stochastic forward passes to obtain observed outputs y (“detection results”) corresponding to input x (“indicator data”). The limitation “multi-time forward propagation on the respective indicator data” includes performing T stochastic forward passes of input data points x. Since the neural network classifies handwritten digits (see page 6, § 5.2, lines 12-18) the input images are indicator data and output classifications are detection results.)
subsequent to obtaining the respective detection results, determining the respective uncertainty of the detection results corresponding to the respective indicator data based on the respective detection results. (Page 4, col. 1, from the line “We estimate the second raw moment” to the second line below equation 8. Equation 8 discloses an uncertainty based on the observed output y (“detection results”).)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have substituted Susaiyah’s neural network model with Gal’s dropout neural network model, and to have computed uncertainty based on Gal’s calculations. A motivation for the combination is that Gal’s calculations better reflects classification uncertainty far from the training data when compared to softmax outputs. (Gal, page 1, col. 2, lines 5-17)
Regarding claim 3, the combination of Susaiyah and Gal teaches: The method according to claim 2,
However, Susaiyah does not explicitly teach: wherein the determining of the respective uncertainty of the detection results subsequent to obtaining the respective detection results comprises: determining at least one of a detection result distribution variance and a detection result distribution standard deviation of the respective detection results corresponding to the multi-time forward propagation; and determining the respective uncertainty of the detection results corresponding to the respective indicator data based on the at least one of the detection result distribution variance and the detection result distribution standard deviation.
But Gal teaches: wherein the determining of the respective uncertainty of the detection results subsequent to obtaining the respective detection results comprises: determining at least one of a detection result distribution variance and a detection result distribution standard deviation of the respective detection results corresponding to the multi-time forward propagation; and (Page 4, col. 1, from the line “We estimate the second raw moment” to equation 7 discloses calculating a detection result variance.)
determining the respective uncertainty of the detection results corresponding to the respective indicator data based on the at least one of the detection result distribution variance and the detection result distribution standard deviation. (Page 4, col. 1, from the line “We estimate the second raw moment” to the second line below equation 8.)
A motivation for the combination is the same as the motivation given for claim 2.
Regarding claim 4, the combination of Susaiyah and Gal teaches: The method according to claim 3, the method further comprising:
However, Susaiyah does not explicitly teach: determining a detection result mean based on the respective detection results corresponding to the multi-time forward propagation; and determining the respective detection results corresponding to the respective indicator data based on the detection result mean.
But Gal teaches: determining a detection result mean based on the respective detection results corresponding to the multi-time forward propagation; and (Page 3, col. 1, lines 17-18 and page 4, col. 1, start of § 4 to line 4 below equation 6 discloses obtaining T stochastic forward passes to obtain an average of observed outputs y (“detection results”) corresponding to input x (“indicator data”). The limitation “multi-time forward propagation” includes performing T stochastic forward passes of input data points x.)
determining the respective detection results corresponding to the respective indicator data based on the detection result mean. (Page 4, equation 6 discloses calculating a mean based on observed outputs y (“detection results”). The observed outputs are identified when calculating the detection result mean.)
A motivation for the combination is the same as the motivation given for claim 2.
Regarding claim 10, Susaiyah teaches: The apparatus of claim 9,
…
the first determining code comprises: second obtaining code configured to cause the at least one first processor to obtain, by the
subsequent to obtaining the respective detection results, second determining code configured to cause the at least one first processor to determine the respective uncertainty of the detection results corresponding to the respective indicator data based on the respective detection results. ([0018], lines 1-7 discloses a set of instructions, which corresponds to second determining code. In [0078]-[0080], the entropy for each instance is calculated based on the posterior probability measures for the two classes A and B.)
However, Susaiyah does not explicitly teach: wherein the deep neural network model is a random dropout neural network model, wherein the random dropout neural network model, during operation, randomly drops inside neuron connections based on a preset dropout rate; and
obtain, by the random dropout neural network model, respective detection results based on multi-time forward propagation on the respective indicator data;
But Gal teaches: wherein the deep neural network model is a random dropout neural network model, wherein the random dropout neural network model, during operation, randomly drops inside neuron connections based on a preset dropout rate; and (Page 3, col. 1, lines 13-20 and lines 1-6 below equation 1. A preset dropout rate is pi)
… obtain, by the random dropout neural network model, respective detection results based on multi-time forward propagation on the respective indicator data; (Page 3, col. 1, lines 17-18 and page 4, col. 1, start of § 4 to line 4 below equation 6 discloses obtaining T stochastic forward passes to obtain observed outputs y (“detection results”) corresponding to input x (“indicator data”). The limitation “multi-time forward propagation on the respective indicator data” includes performing T stochastic forward passes of input data points x. Since the neural network classifies handwritten digits (see page 6, § 5.2, lines 12-18) the input images are indicator data and output classifications are detection results.)
subsequent to obtaining the respective detection results, the detection results corresponding to the respective indicator data based on the respective detection results. (Page 4, col. 1, from the line “We estimate the second raw moment” to the second line below equation 8. Equation 8 discloses an uncertainty based on the observed output y (“detection results”).)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have substituted Susaiyah’s neural network model with Gal’s dropout neural network model, and to have computed uncertainty based on Gal’s calculations. A motivation for the combination is that Gal’s calculations better reflects classification uncertainty far from the training data when compared to softmax outputs. (Gal, page 1, col. 2, lines 5-17)
Regarding claim 11, the combination of Susaiyah and Gal teaches: The apparatus of claim 10,
Susaiyah teaches: wherein the second determining code comprises: third determining code configured to cause the at least one first processor ([0018], lines 1-7 discloses a set of instructions, which corresponds to the second determining code comprising third determining code.)
fourth determining code configured to cause the at least one first processor ([0018], lines 1-7 discloses a set of instructions, which corresponds to the second determining code comprising fourth determining code.)
However, Susaiyah does not explicitly teach: determine at least one of a detection result distribution variance and a detection result distribution standard deviation of the respective detection results corresponding to the multi-time forward propagation; and
determine the respective uncertainty of the detection results corresponding to the respective indicator data based on the at least one of the detection result distribution variance and the detection result distribution standard deviation.
But Gal teaches: determine at least one of a detection result distribution variance and a detection result distribution standard deviation of the respective detection results corresponding to the multi-time forward propagation; and (Page 4, col. 1, from the line “We estimate the second raw moment” to equation 7 discloses calculating a detection result variance.)
determine the respective uncertainty of the detection results corresponding to the respective indicator data based on the at least one of the detection result distribution variance and the detection result distribution standard deviation. (Page 4, col. 1, from the line “We estimate the second raw moment” to the second line below equation 8.)
A motivation for the combination is the same as the motivation given for claim 10.
Regarding claim 12, the combination of Susaiyah and Gal teaches: The apparatus of claim 11,
Susaiyah teaches: wherein the first selecting data comprises: fifth determining code configured to cause the at least one first processor ([0018], lines 1-7 discloses a set of instructions, which corresponds to the first selecting data/code comprising fifth determining code.
sixth determining code configured to cause the at least one first processor ([0018], lines 1-7 discloses a set of instructions, which corresponds to the first selecting data/code comprising sixth determining code.)
However, Susaiyah does not explicitly teach: determine a detection result mean based on the respective detection results corresponding to the multi-time forward propagation; and
determine the respective detection results corresponding to the respective indicator data based on the detection result mean.
But Gal teaches: determine a detection result mean based on the respective detection results corresponding to the multi-time forward propagation; and (Page 3, col. 1, lines 17-18 and page 4, col. 1, start of § 4 to line 4 below equation 6 discloses obtaining T stochastic forward passes to obtain an average of observed outputs y (“detection results”) corresponding to input x (“indicator data”). The limitation “multi-time forward propagation” includes performing T stochastic forward passes of input data points x.)
determine the respective detection results corresponding to the respective indicator data based on the detection result mean. (Page 4, equation 6 discloses calculating a mean based on observed outputs y (“detection results”). The observed outputs are identified when calculating the detection result mean.)
A motivation for the combination is the same as the motivation given for claim 10.
Claims 16-18 each recites an product which implements the same features as the method of claims 2-4, respectively, and are therefore rejected for at least the same reasons.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure: Gal et al. (“Dropout as a Bayesian Approximation: Appendix”) explains variables and formulas used in “Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning” to Gal et al.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Asher H. Jablon whose telephone number is (571)270-7648. The examiner can normally be reached Monday - Friday, 9:00 am - 6:00 pm.
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, Abdullah Al Kawsar can be reached at (571)270-3169. 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.J./Examiner, Art Unit 2127
/ABDULLAH AL KAWSAR/Supervisory Patent Examiner, Art Unit 2127
Prosecution Insights