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 .
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
Claims 1-7 invoke means for interpretation. The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
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 1-6, 12 and 14 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because Claims 1-6 and 14 are directed to software per se. Claims 1-6 contain a means for executing the algorithm, but the specification paragraph 79 says the means may be software. Claim 12 are directed to a signal per se. To overcome the rejection to claims 1-6 and 14 include a structural element like a processor and memory. To overcome the rejection to claim 12 amend to claim “non-transitory computer readable media”.
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.
Claims 10 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.
Claim 10 recites “the means”, but there is insufficient antecedent basis for this element.
Claims 4 and 10 recite “meth the accuracy rule for N consecutive times…”
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 5-9 and 11-14 are rejected under 35 U.S.C. 103 as being unpatentable over US20210117773A1 to Sollami et al and US20190259474A1 to Wang et al.
Claim 3 rejected under 35 U.S.C. 103 as being unpatentable over US20210117773A1 to Sollami et al, US20190259474A1 to Wang et al and Generative Adversarial Nets by Goodfellow et al.
Claims 4 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over US20210117773A1 to Sollami et al, US20190259474A1 to Wang et al and FeaSel-Net: A Recursive Feature Selection Callback in Neural Networks by Fischer et al.
Sollami teaches claims 1 and 8. An apparatus comprising means for performing:
initializing a first set of trainable parameters for a first machine learning based model outputting synthetic data; and (Sollami para 25 “discriminator network and generator network may be trained…” Generator is the first model.)
initializing a second set of trainable parameters for a second machine learning based model classifying input data to synthetic data or real data outputting feedback, wherein the first machine learning based model and the second machine learning based model are competing models; (Sollami para 25 “discriminator network and generator network may be trained…” The discriminator and generator are part of a “Generative adversarial network (GAN)” so they are competing. Sollami abs. Discriminator is the second model.)
determining, whether an end criterium is met; (Sollami para 25 “a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.”)
performing a first training process comprising:
obtaining real samples; (Sollami para 72 “At 1002, a catalog may be received. … Each catalog entry may include images of an item, and a description, such as text, and metadata, about the item.”)
inputting the real samples and synthetic samples output by the first machine learning based model to the second machine learning based model to train the second set of trainable parameters; (Sollami abs “The GANs may generate images including images of generated items, which may be replaced with images of items from the catalog entries to create feature model training images.” Creating feature model training images is using the synthetic/generated images to train the second discriminator model.)
determining accuracy of the second machine learning based model;
applying a preset accuracy rule to determine whether the accuracy of the second machine learning based model meets the accuracy rule; and (Sollami para 25 “The GANs may generate images including images of generated items, which may be replaced with images of items from the catalog entries to create feature model training images.”)
as long as the end criterium and the accuracy rule are not met, repeating inputting to the second machine learning based model the feedback from the second machine learning based model to retrain the second set of trainable parameters by re-using the samples, determining accuracy and applying the preset accuracy rule; (Sollami para 25 “he discriminator network and generator network may be trained until the discriminator network reaches a threshold level of accuracy on the training data set and a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.” Sollami para 23 “backpropagation may be used to adjust the weights of the discriminator network, training the discriminator network based on errors made by the discriminator network.” Since Sollami does not teach generating new images at each training iteration, see fig. 10 1018, Sollami is re-using the samples.)
performing, after the first training process, when the accuracy rule is met but the end criterium is not met, a second training process comprising: (Sollami para 24 “After the discriminator network of the GAN has been trained for a set length of time or on a set number of images from the GAN training data set, a generator network of the GAN may be trained.” The Discriminator is the second model and it is trained when the accuracy rule (threshold percentage of positive examples) is met.)
inputting feedback from the second machine learning based model and random noise to the first machine learning model to train the first set of trainable parameters and to obtain new synthetic samples as output of the first machine learning model; (Sollami para 24 “The image output by the generator network may be input to the discriminator network, which may output an indication of whether the image is a positive example, … When the discriminator network indicates that the image is not a positive example, for example, is a negative example, the weights of the generator network may be adjusted, for example, through backpropagation, training the generator network…. The generator network may be trained for any suitable length of time, using any suitable number of random inputs.”)
repeating, as long as the end criterium is not met, performing the first training process and the second training process; and (Sollami para 25 “Training may alternate between the discriminator network and the generator network, and may continue for any suitable period of time. For example, the discriminator network and generator network may be trained until the discriminator network reaches a threshold level of accuracy on the training data set and a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.”)
Sollami doesn’t teach storing the model.
However, Wang teaches storing, after determining that the end criterium is met, at least the first machine learning model trained. (Wang para 83 “FIG. 5A. At 224, trained models for discriminator 226 and generator 228 may be stored.”)
Wang, Sollami and the claims are all GANs. It would have been obvious to a person having ordinary skill in the art, at the time of filing, to store the trained model because the model has to be stored if it’s going to be used later. And training a model and then throwing it away is wasteful.
Sollami teaches claims 2 and 9. The apparatus of claim 1, wherein the preset accuracy rule is met at least when the accuracy is above a preset threshold. (Sollami para 25 “he discriminator network and generator network may be trained until the discriminator network reaches a threshold level of accuracy on the training data set and a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.” The threshold percentage is the preset threshold.)
Sollami teaches claim 3. The apparatus of claim 1, wherein the means are further configured to perform maintaining a value of a counter based at least on how many times the first training process and the second training process are performed, and the end criterium is based on (Sollami para 57 “A training cycle for the GANs 140 may alternate between training cycles for the discriminator networks and the generator networks any suitable number of times, and the end of the training cycle for the GANs 140 may be determined in any suitable manner.”)
Sollami doesn’t teach a limit for the number of times/count.
However, Goodfellow teaches the end criterium is based on a preset limit for the value of the counter. (Goodfellow algorithm 1 on page 4 optimizes the discriminator (second model) k times, and optimizes the generator (first model) for a number of training iterations. The number of training iterations is the preset limit.)
Sollami, Goodfellow and the claims are all GANs. It would have been obvious to a person having ordinary skill in the art, at the time of filing, to use an iteration count stop to end training where gradient descent can’t find a minimum error.
Sollami teaches claims 4 and 10. The apparatus of claim 1, wherein the means are configured to determine that the end criterium is met when the first training process has meth the accuracy rule
Sollami doesn’t teach a limit for the count.
However, Fischer teaches to determine that the end criterium is met when the first training process has meth the accuracy rule for N consecutive times, wherein the N is a positive integer having a value bigger than one. (Fischer sec. 3.2.2 “Threshold criterion: The loss gradient or accuracy values surpass a pre-defined threshold τ. (b) Consistency criterion: The threshold is surpassed for a minimal number of consecutive epochs… epoch e=40…”)
Sollami, Fischer and the claims all train for accuracy. It would have been obvious to a person having ordinary skill in the art, at the time of filing, to train for a consecutive number of accuracy threshold meets in order to ensure consistently high accuracy.
Sollami teaches claims 5 and 11. The apparatus of claim 1, wherein the means are further configured to determine the accuracy by comparing sample by sample correctness of the classification of the second machine learning model. (Sollami para 25 “he discriminator network and generator network may be trained until the discriminator network reaches a threshold level of accuracy on the training data set and a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.” The positive example determination is a sample by sample correctness of the classification of the second model.)
Sollami teaches claim 6. The apparatus of claim 1, wherein the first machine learning based model and the second machine learning based model are based on generative adversarial networks. (Sollami abs “Generative adversarial network (GAN) training data sets may be generated from the images of the items, the additional images, and the categories. GANs may be trained with the GAN training data sets.”)
Sollami teaches claim 7. The apparatus of claim 1, the apparatus comprising at least one processor, and at least one memory including computer program code, wherein the at least one processor with the at least one memory and computer program code provide said means. (Sollami figs. 1-5)
Sollami teaches claims 12 and 14. A computer readable medium comprising program instructions stored thereon for at least one of a first functionality or a second functionality, for performing corresponding functionality, (Sollami fig. 10 and fig. 1-5)
wherein the first functionality comprises at least following:
initializing a first set of trainable parameters for a first machine learning based model outputting synthetic data; (Sollami para 25 “discriminator network and generator network may be trained…” Generator is the first model.)
initializing a second set of trainable parameters for a second machine learning based model classifying input data to synthetic data or real data outputting feedback, wherein the first machine learning based model and the second machine learning based model are competing models; (Sollami para 25 “discriminator network and generator network may be trained…” The discriminator and generator are part of a “Generative adversarial network (GAN)” so they are competing. Sollami abs. Discriminator is the second model.)
determining, whether an end criterium is met; (Sollami para 25 “a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.”)
performing a first training process comprising:
obtaining real samples; (Sollami para 72 “At 1002, a catalog may be received. … Each catalog entry may include images of an item, and a description, such as text, and metadata, about the item.”)
inputting the real samples and synthetic samples output by the first machine learning based model to the second machine learning based model to train the second set of trainable parameters; (Sollami abs “The GANs may generate images including images of generated items, which may be replaced with images of items from the catalog entries to create feature model training images.” Creating feature model training images is using the synthetic/generated images to train the second discriminator model.)
determining accuracy of the second machine learning based model;
applying a preset accuracy rule to determine whether the accuracy of the second machine learning based model meets the accuracy rule; and (Sollami para 25 “The GANs may generate images including images of generated items, which may be replaced with images of items from the catalog entries to create feature model training images.”)
as long as the end criterium and the accuracy rule are not met, repeating inputting to the second machine learning based model the feedback from the second machine learning based model to retrain the second set of trainable parameters by reusing the samples, determining accuracy and applying the preset accuracy rule; (Sollami para 25 “he discriminator network and generator network may be trained until the discriminator network reaches a threshold level of accuracy on the training data set and a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.” Sollami para 23 “backpropagation may be used to adjust the weights of the discriminator network, training the discriminator network based on errors made by the discriminator network.” Since Sollami does not teach generating new images at each training iteration, see fig. 10 1018, Sollami is re-using the samples.)
performing, after the first training process, when the accuracy rule is met but the end criterium is not met, a second training process comprising inputting feedback from the second machine learning based model and random noise to the first machine learning model to train the first set of trainable parameters and to obtain new synthetic samples as output of the first machine learning model; (Sollami para 24 “The image output by the generator network may be input to the discriminator network, which may output an indication of whether the image is a positive example, … When the discriminator network indicates that the image is not a positive example, for example, is a negative example, the weights of the generator network may be adjusted, for example, through backpropagation, training the generator network…. The generator network may be trained for any suitable length of time, using any suitable number of random inputs.”)
repeating, as long as the end criterium is not met, performing the first training process and the second training process; and (Sollami para 25 “Training may alternate between the discriminator network and the generator network, and may continue for any suitable period of time. For example, the discriminator network and generator network may be trained until the discriminator network reaches a threshold level of accuracy on the training data set and a threshold percentage of images output by the generator network are indicated by the discriminator network as being images of positive examples.”)
wherein the second functionality comprises at least following:
obtaining one or more sets of real data;
obtaining one or more sets of synthetic data by inputting noise to the first machine learning model trained using the first functionality; and
training a machine learning based classifier using both the real data and the synthetic data. (Sollami para 24 “The image output by the generator network may be input to the discriminator network, which may output an indication of whether the image is a positive example, … When the discriminator network indicates that the image is not a positive example, for example, is a negative example, the weights of the generator network may be adjusted, for example, through backpropagation, training the generator network…. The generator network may be trained for any suitable length of time, using any suitable number of random inputs.”)
Sollami doesn’t teach storing the model.
However, Wang teaches storing, after determining that the end criterium is met, at least the first machine learning model trained. (Wang para 83 “FIG. 5A. At 224, trained models for discriminator 226 and generator 228 may be stored.”)
Wang, Sollami and the claims are all GANs. It would have been obvious to a person having ordinary skill in the art, at the time of filing, to store the trained model because the model has to be stored if it’s going to be used later. And training a model and then throwing it away is wasteful.
Sollami teaches claim 13. The computer readable medium according of claim 12, wherein the computer readable medium is a non-transitory computer readable medium. (Sollami figs. 1-5.)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Austin Hicks whose telephone number is (571)270-3377. The examiner can normally be reached Monday - Thursday 8-4 PST.
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, Mariela Reyes can be reached at (571) 270-1006. 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.
/AUSTIN HICKS/ Primary Examiner, Art Unit 2142
Prosecution Insights