DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . This action is in response to an action filed on February 27th, 2026. Claims 1, and 3-16 are pending in the current application. Claims 1, 3, 5, 11, 15 and 16 have been amended and claim 2 has been canceled.
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, 15, and 16 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.
The claims are generally narrative and indefinite, failing to conform with current U.S. practice. They are replete with grammatical and idiomatic errors, such as “obtaining first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, “obtaining second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data”, and “wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model.”
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.
Claim(s) 1, and 3-16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding claim 1, Under Step 1 of the Subject Matter Eligibility Test of Products and Processes, the claim is directed towards a machine, which is one of the four statutory categories.
Next, under a Step 2A Prong 1 Analysis, the claim recites the following limitations which are interpreted to be, under the broadest reasonable interpretation, groupings of abstract ideas.
determine whether each of a plurality of pieces of first data satisfies a first condition relating to labeling; (mental process)
create a first learning model capable of the labeling based on a first group being a group of pieces of the first data which satisfy the first condition and which are labeled (mental process)
convert a second group being a group of pieces of the first data which do not satisfy the first condition and which are not labeled so that a distribution of the second group is close to a distribution of the first group (mental process)
execute the labeling for the second group based on the first learning model and the second group (mental process)
create, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling (mental process)
Therefore, we have to examine the claim under Step 2A prong 2, which considers the additional elements within the claim. The claims additional elements are:
at least one processor
obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data;
obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data;
wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model;
by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model.
The limitations, “at least one processor”, and “wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model;” as drafted, are considered to be mere instructions to apply an exception, as it instructs to use the processor and first machine learning model to perform the abstract idea. (See MPEP 2106.05(f)) The limitations “obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, and “obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data” are considered to be insignificant extra-solution activity. (See MPEP 2106.05(g)) Lastly, the limitation “by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model” is given negligible patentable weight, as the second learning model only needs to be capable of performing these steps. Even if the limitation was to be positively recited, it would only amount to “generally linking” the limitations to the second machine learning model. (See MPEP 2106.06(h)) Therefore, the additional elements do not integrate the abstract idea into a practical application. The claim is directed to an abstract idea.
Under a Step 2B analysis, the claim's addition elements do not amount to significantly
more than the judicial exception as explained above in Step 2A prong 2. Additionally, to “obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, and “obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data” is considered to be well-understood, routine and conventional, as it is considered to be receiving or transmitting data over a network. (See MPEP 2106.05(d)(ii)) Therefore, the claim is ineligible.
Regarding claim 15, Under Step 1 of the Subject Matter Eligibility Test of Products and Processes, the claim is directed towards a process, which is one of the four statutory categories.
Next, under a Step 2A Prong 1 Analysis, the claim recites the following limitations which are interpreted to be, under the broadest reasonable interpretation, groupings of abstract ideas.
determining whether each of a plurality of pieces of first data satisfies a first condition relating to labeling; (mental process)
creating a first learning model capable of the labeling based on a first group being a group of pieces of the first data which satisfy the first condition and which are labeled (mental process)
converting a second group being a group of pieces of the first data which do not satisfy the first condition and which are not labeled so that a distribution of the second group is close to a distribution of the first group (mental process)
executing the labeling for the second group based on the first learning model and the second group (mental process)
creating, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling (mental process)
Therefore, we have to examine the claim under Step 2A prong 2, which considers the additional elements within the claim. The claims additional elements are:
obtaining first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data;
obtaining second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data;
wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model;
by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model
The limitations, “wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model;” as drafted, are considered to be mere instructions to apply an exception, as it instructs to use the first machine learning model to perform the abstract idea. (See MPEP 2106.05(f)) The limitations “obtaining first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, and “obtaining second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data” are considered to be insignificant extra-solution activity. (See MPEP 2106.05(g)) Lastly, the limitation “by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model” is given negligible patentable weight, as the second learning model only needs to be capable of performing these steps. Even if the limitation was to be positively recited, it would only amount to “generally linking” the limitations to the second machine learning model. (See MPEP 2106.06(h)) Therefore, the additional elements do not integrate the abstract idea into a practical application. The claim is directed to an abstract idea.
Under a Step 2B analysis, the claim's addition elements do not amount to significantly
more than the judicial exception as explained above in Step 2A prong 2. Additionally, “obtaining first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, and “obtaining second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data” is considered to be well-understood, routine and conventional, as it is considered to be receiving or transmitting data over a network. (See MPEP 2106.05(d)(ii)) Therefore, the claim is ineligible.
Regarding claim 16, Under Step 1 of the Subject Matter Eligibility Test of Products and Processes, the claim is directed towards a manufacture, which is one of the four statutory categories.
Next, under a Step 2A Prong 1 Analysis, the claim recites the following limitations which are interpreted to be, under the broadest reasonable interpretation, groupings of abstract ideas.
determine whether each of a plurality of pieces of first data satisfies a first condition relating to labeling; (mental process)
create a first learning model capable of the labeling based on a first group being a group of pieces of the first data which satisfy the first condition and which are labeled (mental process)
convert a second group being a group of pieces of the first data which do not satisfy the first condition and which are not labeled so that a distribution of the second group is close to a distribution of the first group (mental process)
and execute the labeling for the second group based on the first learning model and the second group (mental process)
create, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling. (mental process)
Therefore, we have to examine the claim under Step 2A prong 2, which considers the additional elements within the claim. The claims additional elements are:
A non-transitory computer-readable information storage medium
obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data;
obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data;
wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model;
by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model
The limitations, “a non-transitory computer-readable information storage medium”, and “wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model;” as drafted, are considered to be mere instructions to apply an exception, as it instructs to use the medium and first machine learning model to perform the abstract idea. (See MPEP 2106.05(f)) The limitations “obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, and “obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data” are considered to be insignificant extra-solution activity. (See MPEP 2106.05(g)) Lastly, the limitation “by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model” is given negligible patentable weight, as the second learning model only needs to be capable of performing these steps. Even if the limitation was to be positively recited, it would only amount to “generally linking” the limitations to the second machine learning model. (See MPEP 2106.06(h)) Therefore, the additional elements do not integrate the abstract idea into a practical application. The claim is directed to an abstract idea.
Under a Step 2B analysis, the claim's addition elements do not amount to significantly
more than the judicial exception as explained above in Step 2A prong 2. Additionally, to “obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data”, and “obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data” is considered to be well-understood, routine and conventional, as it is considered to be receiving or transmitting data over a network. (See MPEP 2106.05(d)(ii)) Therefore, the claim is ineligible.
Regarding claim 3, the claim recites to “create, based on the second learning model, a second condition which is different from the first condition and which relates to the labeling;” and “determine whether each of a plurality of pieces of second data different from the plurality of pieces of first data satisfies the second condition” The limitations, as drafted, are considered to be, under the broadest reasonable interpretation, “mental processes” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 1.
Regarding claim 4, the claim recites to “create a third learning model capable of the labeling based on a third group being a group of pieces of the second data which satisfy the second condition and which are labeled;”, “convert a fourth group being a group of pieces of the second data which do not satisfy the second condition and which are not labeled so that a distribution of the fourth group is close to a distribution of the third group;”, and “execute the labeling for the fourth group based on the third learning model and the fourth group.” The limitations, as drafted, are considered to be, under the broadest reasonable interpretation, “mental processes” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 3.
Regarding claim 5, the claim recites to “execute, based on the second learning model, the labeling for each of a plurality of pieces of second data different from the plurality of pieces of first data.” The limitation, as drafted, is considered to be, under the broadest reasonable interpretation, a “mental process” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 1.
Regarding claim 6, the claim recites to “create a third learning model capable of the labeling based on a third group being a group of pieces of the second data labeled by the second learning model;”, “convert a fourth group being a group of pieces of the second data which are not labeled by the second learning model so that a distribution of the fourth group is close to a distribution of the third group;”, and “execute the labeling for the fourth group based on the third learning model and the fourth group.” The limitations, as drafted, are considered to be, under the broadest reasonable interpretation, “mental processes” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 5.
Regarding claim 7, the claim recites to “create, based on the first group, the third group, and the fourth group a fourth learning model which is different from any one of the first learning model, the second learning model, or the third learning model and which is capable of the labeling.” The limitation, as drafted, is considered to be, under the broadest reasonable interpretation, a “mental process” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 4.
Regarding claim 8, the claim recites to “determine, based on similarity between the distribution of the first group and the distribution of the third group, whether to use the first group to create the fourth learning model”, “create the fourth learning model without based on the first group when to use the first group”, and “to create the fourth learning model based on the first group when to use the first group.” The limitations, as drafted, are considered to be, under the broadest reasonable interpretation, “mental processes” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 7.
Regarding claim 9, the claim recites to “create the fourth learning model further based on the second group” The limitation, as drafted, is considered to be, under the broadest reasonable interpretation, a “mental process” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 7.
Regarding claim 10, the claim recites to “determine, based on similarity between the distribution of the second group and the distribution of the fourth group, whether to use the second group to create the fourth learning model”, “create the fourth learning model without based on the second group when to use the second group”, and to “create the fourth learning model based on the second group when to use the second group.” The limitations, as drafted, are considered to be, under the broadest reasonable interpretation, “mental processes” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 9.
Regarding claim 11, the claim recites to “create the second learning model based on the second group and before being converted.” The limitation, as drafted, is considered to be, under the broadest reasonable interpretation, a “mental process” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 1.
Regarding claim 12, the claim recites to “execute, based on the second group, additional learning for the first learning model which has learned the first group.” The limitation, as drafted, is considered to be, under the broadest reasonable interpretation, a “mental process” which is a grouping of abstract idea. Therefore, the claim is rejected on the same basis as claim 1.
Regarding claim 13, the claim recites to “provide the pieces of the first data satisfying the first condition to an administrator who executes the labeling;”, “receive specification for the label by the administrator”, and “execute the labeling for the first group based on the specification by the administrator.” To “execute the labeling for the first group based on the specification by the administrator” is considered to be a “mental process”, which is a grouping of abstract idea, with the additional elements to “provide the pieces of the first data satisfying the first condition to an administrator who executes the labeling;”, and “receive specification for the label by the administrator” is considered to be insignificant extra-solution activity, (See MPEP 2106.05(g)) and considered to be well-understood, routine, and conventional, as it is considered to be sending or receiving data over a network. (See MPEP 2106.05(d)(ii)) Therefore, the claim is rejected on the same basis as claim 1.
Regarding claim 14, the claim recites “each of the plurality of pieces of first data indicates an action of a user who uses a predetermined service”, “the predetermined service is provided based on user information on the user”, “the labeling is processing of determining whether the action of the user having valid user information is fraudulent”, and “the label is a determined fraud label indicating that the fraud is determined.” The limitations, as drafted, merely indicates the field of use or technological environment, and “generally links” the abstract idea to determining fraud. (See MPEP 2106.05(h)) Therefore, the claim is rejected on the same basis as claim 1.
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.
Claim(s) 1, 3-12, 15 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Gan et al. (Herein referred to as Gan) (U.S. Patent No. US 11664820 B2) in view of Jian Liang. (Herein referred to as Liang) (Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfer)
Regarding claim 1, Gan teaches a learning system, comprising at least one processor (“this specification describes an apparatus comprising: at least one processor; and at least one memory including computer program code”, column 5, lines 8-10) configured to: determine whether each of a plurality of pieces of first data satisfies a first condition relating to labeling; (“the goal is to train a classifier for each of these target domains. As we do not have labelled data from the target domains, we will need to perform domain adaptation.” Column 14, lines 48-50) (Whether or not the piece of data in this particular domain acts as a first condition.) create a first learning model capable of the labeling based on a first group being a group of pieces of the first data which satisfy the first condition and which are labeled; (“we have a labelled source domain S in the system, and we have a classifier (M.sub.S) trained using supervised learning algorithms for S.” Column 14, lines 44-46) (The classifier acts as our first learning model with the source domain acting as our first data which are labeled.) convert a second group being a group of pieces of the first data which do not satisfy the first condition and which are not labeled so that a distribution of the second group is close to a distribution of the first group; (“Let us say that T1 is introduced in the system. Currently, we only have S already in the system (e.g. the initial source encoder 91) along with its pre-trained inference model M.sub.S. As such, we perform a domain adaptation step between S and T1, which results in a model M.sub.T1 for the domain T1 (e.g. according to the one or more processes 50, 60 or 70 in FIG. 5, 6 or 7).”, Column 14, line 51-57; See also Column 14, lines 60-63) (If the data does not satisfy a domain, a computation is made to determine whether the data diverges smaller to source data or an existing domain, and the data is converted for domain adaptation.) and execute the labeling for the second group based on the first learning model and the second group; (“we perform a domain adaptation step between S and T1, which results in a model M.sub.T1 for the domain T1 (e.g. according to the one or more processes 50, 60 or 70 in FIG. 5, 6 or 7).”, Column 14, line 54-57;) (Domain adaptation is performed, and a new classifier model is output to label data pertaining to the adapted domain.)
However, Gan does not explicitly teach to obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data; obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data; wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model; and create, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model.
Liang teaches obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data; (“For a vanilla UDA task, we are given ns labeled samples {xi s, yi s}ns i=1 from the source domain Ds where xi s ∈ Xs, yi s ∈ Ys, and also nt unlabeled samples {xi t}nt i=1 from the target domain Dt where xi t ∈ Xt. The goal of UDA is to predict the labels {yi t}nt i=1 in the target domain, where yi t ∈ Yt, and the source task Xs → Ys is assumed to be the same with the target task Xt → Yt.”, pg. 4, left column, second paragraph) (The labeled samples correspond to training data comprising label-data pairs which belong to a source domain, which corresponds to a first group. The task is “assumed to be the same with the target task”, corresponding to labels added to target data.) obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data; wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model; (“First, we train the classification model, consisting of a feature encoding module and a hypothesis module, from the source data and then transfer the source model to the target domain without accessing the source data.”, pg. 4, left column, third paragraph) (A classifier is trained from the source data and predicts labels from the source data, corresponding to a second training data comprising label-data pairs, the labels corresponding to the target group/data. (second group)) and create, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model. (“Then, we present a novel framework, Source HypOthesis Transfer (SHOT), to learn the target-specific feature encoding module using self-supervised learning and semi-supervised learning, with the source hypothesis fixed. Finally, using the predictions for the target domain, we further employ a semi-supervised learning algorithm to enforce labeling information propagation from confidently labeled target samples to the remaining target samples with low confidences.”, pg. 4, left column, third paragraph) (Hypothesis transfer is utilized to create a second model.) (This limitation is interpreted very broadly and has negligible patentable weight as currently drafted, as the broadest reasonable interpretation is a created model capable of labeling and adjusting parametesr when conditions are met. The framework of the model is shown to be capable of labeling by adjusting the parameters when conditions are met, teaching this limitation.)
Therefore, it would have been considered obvious to one of ordinary skill in the art, prior to the filing date of the current application, to combine the domain adaptation of Gan with the transferring process of Liang. One would have been motivated to combine the teaching, prior to the application’s filing date, as this allows one to “maximize the mutual information between intermediate feature representations and outputs of the classifier”, as disclosed in Liang. (“We are then motivated to make SHOT adapt the feature encoding mod ule by fine-tuning the source feature encoding module while freezing the source hypothesis, to maximize the mutual information between intermediate feature representations and outputs of the classifier, since information maximization [26], [27] can encourage the classifier to assign disparate one-hot outputs to different target feature representations.”, pg. 2, left column, first paragraph)
Regarding claim 15, Gan teaches A learning method, comprising: determining whether each of a plurality of pieces of first data satisfies a first condition relating to labeling; (“the goal is to train a classifier for each of these target domains. As we do not have labelled data from the target domains, we will need to perform domain adaptation.” Column 14, lines 48-50) (Whether or not the piece of data in this particular domain acts as a first condition.) creating a first learning model capable of the labeling based on a first group being a group of pieces of the first data which satisfy the first condition and which are labeled; (“we have a labelled source domain S in the system, and we have a classifier (M.sub.S) trained using supervised learning algorithms for S.” Column 14, lines 44-46) (The classifier acts as our first learning model with the source domain acting as our first data which are labeled.) converting a second group being a group of pieces of the first data which do not satisfy the first condition and which are not labeled so that a distribution of the second group is close to a distribution of the first group; (“Let us say that T1 is introduced in the system. Currently, we only have S already in the system (e.g. the initial source encoder 91) along with its pre-trained inference model M.sub.S. As such, we perform a domain adaptation step between S and T1, which results in a model M.sub.T1 for the domain T1 (e.g. according to the one or more processes 50, 60 or 70 in FIG. 5, 6 or 7).”, Column 14, line 51-57; See also Column 14, lines 60-63) (If the data does not satisfy a domain, a computation is made to determine whether the data diverges smaller to source data or an existing domain, and the data is converted for domain adaptation.) and executing the labeling for the second group based on the first learning model and the second group; (“we perform a domain adaptation step between S and T1, which results in a model M.sub.T1 for the domain T1 (e.g. according to the one or more processes 50, 60 or 70 in FIG. 5, 6 or 7).”, Column 14, line 54-57;) (Domain adaptation is performed, and a new classifier model is output to label data pertaining to the adapted domain.)
However, Gan does not explicitly teach to obtaining first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data; obtaining second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data; wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model; and creating, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model.
Liang teaches obtaining first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data; (“For a vanilla UDA task, we are given ns labeled samples {xi s, yi s}ns i=1 from the source domain Ds where xi s ∈ Xs, yi s ∈ Ys, and also nt unlabeled samples {xi t}nt i=1 from the target domain Dt where xi t ∈ Xt. The goal of UDA is to predict the labels {yi t}nt i=1 in the target domain, where yi t ∈ Yt, and the source task Xs → Ys is assumed to be the same with the target task Xt → Yt.”, pg. 4, left column, second paragraph) (The labeled samples correspond to training data comprising label-data pairs which belong to a source domain, which corresponds to a first group. The task is “assumed to be the same with the target task,” corresponding to labels added to target data.) obtaining second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data; wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model; (“First, we train the classification model, consisting of a feature encoding module and a hypothesis module, from the source data and then transfer the source model to the target domain without accessing the source data.”, pg. 4, left column, third paragraph) (A classifier is trained from the source data and predicts labels from the source data, corresponding to a second training data comprising label-data pairs, the labels corresponding to the target group/data. (second group)) and creating, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model. (“Then, we present a novel framework, Source HypOthesis Transfer (SHOT), to learn the target-specific feature encoding module using self-supervised learning and semi-supervised learning, with the source hypothesis fixed. Finally, using the predictions for the target domain, we further employ a semi-supervised learning algorithm to enforce labeling information propagation from confidently labeled target samples to the remaining target samples with low confidences.”, pg. 4, left column, third paragraph) (Hypothesis transfer is utilized to create a second model.) (This limitation is interpreted very broadly and has negligible patentable weight as currently drafted, as the broadest reasonable interpretation is a created model capable of labeling and adjusting parameters when conditions are met. The framework of the model is shown to be capable of labeling by adjusting the parameters when conditions are met, teaching this limitation.)
Therefore, it would have been considered obvious to one of ordinary skill in the art, prior to the filing date of the current application, to combine the domain adaptation of Gan with the transferring process of Liang. One would have been motivated to combine the teaching, prior to the application’s filing date, as this allows one to “maximize the mutual information between intermediate feature representations and outputs of the classifier”, as disclosed in Liang. (“We are then motivated to make SHOT adapt the feature encoding mod ule by fine-tuning the source feature encoding module while freezing the source hypothesis, to maximize the mutual information between intermediate feature representations and outputs of the classifier, since information maximization [26], [27] can encourage the classifier to assign disparate one-hot outputs to different target feature representations.”, pg. 2, left column, first paragraph)
Regarding claim 16, Gan teaches A non-transitory computer-readable information storage medium for storing a program (“…this specification describes a computer-readable medium (such as a non-transitory computer-readable medium) comprising program instructions stored thereon”, Column 4, lines 56-59) for causing a computer to: determine whether each of a plurality of pieces of first data satisfies a first condition relating to labeling; (“the goal is to train a classifier for each of these target domains. As we do not have labelled data from the target domains, we will need to perform domain adaptation.” Column 14, lines 48-50) (The system is capable of determining if the data set does not contain data from a target domain.) create a first learning model capable of the labeling based on a first group being a group of pieces of the first data which satisfy the first condition and which are labeled; (“we have a labelled source domain S in the system, and we have a classifier (M.sub.S) trained using supervised learning algorithms for S.” Column 14, lines 44-46) (The classifier acts as our first learning model with the source domain acting as our first data which are labeled.) convert a second group being a group of pieces of the first data which do not satisfy the first condition and which are not labeled so that a distribution of the second group is close to a distribution of the first group; (“Let us say that T1 is introduced in the system. Currently, we only have S already in the system (e.g. the initial source encoder 91) along with its pre-trained inference model M.sub.S. As such, we perform a domain adaptation step between S and T1, which results in a model M.sub.T1 for the domain T1 (e.g. according to the one or more processes 50, 60 or 70 in FIG. 5, 6 or 7).”, Column 14, line 51-57; See also Column 14, lines 60-63) (If the data does not satisfy a domain, a computation is made to determine whether the data diverges smaller to source data or an existing domain, and the data is converted for domain adaptation.) and execute the labeling for the second group based on the first learning model and the second group; (“we perform a domain adaptation step between S and T1, which results in a model M.sub.T1 for the domain T1 (e.g. according to the one or more processes 50, 60 or 70 in FIG. 5, 6 or 7).”, Column 14, line 54-57;) (Domain adaptation is performed, and a new classifier model is output to label data pertaining to the adapted domain.)
However, Gan does not explicitly teach to obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data; obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data; wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model; and create, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model.
Liang teaches obtain first training data that is pairs each formed of the first data belonging to the first group and a label added to this target data; (“For a vanilla UDA task, we are given ns labeled samples {xi s, yi s}ns i=1 from the source domain Ds where xi s ∈ Xs, yi s ∈ Ys, and also nt unlabeled samples {xi t}nt i=1 from the target domain Dt where xi t ∈ Xt. The goal of UDA is to predict the labels {yi t}nt i=1 in the target domain, where yi t ∈ Yt, and the source task Xs → Ys is assumed to be the same with the target task Xt → Yt.”, pg. 4, left column, second paragraph) (The labeled samples correspond to training data comprising label-data pairs which belong to a source domain, which corresponds to a first group. The task is “assumed to be the same with the target task,” corresponding to labels added to target data.) obtain second training data that is pairs each formed of the first data belonging to the second group and a label added to this target data; wherein the label added to the target data of the second training data is obtained from the labeling by first machine learning model; (“First, we train the classification model, consisting of a feature encoding module and a hypothesis module, from the source data and then transfer the source model to the target domain without accessing the source data.”, pg. 4, left column, third paragraph) (A classifier is trained from the source data and predicts labels from the source data, corresponding to a second training data comprising label-data pairs, the labels corresponding to the target group/data. (second group)) and create, based on the first group and the second group, a second learning model which is different from the first learning model and which is capable of the labeling by (1) adjusting, when the target data of the first training data is input to the second learning model, parameters of the second learning model so that the label of the first training data is output from the second learning model and (2) adjusting, when the target data of the second training data is input to the second learning model, parameters of the second learning model so that the label of the second training data is output from the second learning model. (“Then, we present a novel framework, Source HypOthesis Transfer (SHOT), to learn the target-specific feature encoding module using self-supervised learning and semi-supervised learning, with the source hypothesis fixed. Finally, using the predictions for the target domain, we further employ a semi-supervised learning algorithm to enforce labeling information propagation from confidently labeled target samples to the remaining target samples with low confidences.”, pg. 4, left column, third paragraph) (Hypothesis transfer is utilized to create a second model.) (This limitation is interpreted very broadly and has negligible patentable weight as currently drafted, as the broadest reasonable interpretation is a created model capable of labeling and adjusting parameters when conditions are met. The framework of the model is shown to be capable of labeling by adjusting the parameters when conditions are met, teaching this limitation.)
Therefore, it would have been considered obvious to one of ordinary skill in the art, prior to the filing date of the current application, to combine the domain adaptation of Gan with the transferring process of Liang. One would have been motivated to combine the teaching, prior to the application’s filing date, as this allows one to “maximize the mutual information between intermediate feature representations and outputs of the classifier”, as disclosed in Liang. (“We are then motivated to make SHOT adapt the feature encoding mod ule by fine-tuning the source feature encoding module while freezing the source hypothesis, to maximize the mutual information between intermediate feature representations and outputs of the classifier, since information maximization [26], [27] can encourage the classifier to assign disparate one-hot outputs to different target feature representations.”, pg. 2, left column, first paragraph)
Regarding claim 3, Gan, as modified by Liang, teaches the learning system according to claim 1, wherein the at least one processor is configured to: create, based on the second learning model, a second condition which is different from the first condition and which relates to the labeling; (“the goal is to train a classifier for each of these target domains. As we do not have labelled data from the target domains, we will need to perform domain adaptation.” Column 14, lines 48-50 (Gan)) (The second condition can be data labelled that fits the target domain) and determine whether each of a plurality of pieces of second data different from the plurality of pieces of first data satisfies the second condition. (“the goal is to train a classifier for each of these target domains. As we do not have labelled data from the target domains, we will need to perform domain adaptation.” Column 14, lines 48-50 (Gan)) (For the method to perform domain adaptation, they have to have a way to determine whether the labeled data satisfies a condition.)
Regarding claim 4, Gan, as modified by Liang, teaches the learning system according to claim 3, wherein the at least one processor is configured to: create a third learning model capable of the labeling based on a third group being a group of pieces of the second data which satisfy the second condition and which are labeled; (“we perform a domain adaptation step between T1 and T2, wherein we adapt the weights of M.sub.T1 to output a model for T2, namely M.sub.T2.” Column 15, lines 11-13; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The process that creates the first and second model can be repeated to create a third model.) convert a fourth group being a group of pieces of the second data which do not satisfy the second condition and which are not labeled so that a distribution of the fourth group is close to a distribution of the third group; (“the control module 96 uses a selection technique based on computing domain divergence, i.e. how far are two domains from each other. In particular, the control module 96 may determine a pairwise divergence (T.sub.2, S) or (T.sub.2, T.sub.1) between the target domain (i e T2) and the candidate domains (S and T1), and identify the pair with the smallest divergence.” Column 14, lines 64-67, and Column 15, lines 1-3; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The steps can be repeated to convert pieces of data that do not fit in the first, second, or third domain into a fourth group.) and execute the labeling for the fourth group based on the third learning model and the fourth group (“we perform a domain adaptation step between T1 and T2, wherein we adapt the weights of M.sub.T1 to output a model for T2,” Column 15, lines 11-13; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (With the repeatable steps, a new model is made to label the fourth group.)
Regarding claim 5, Gan, as modified by Liang, teaches the learning system according to claim 1, wherein the at least one processor is configured to execute, based on the second learning model, the labeling for each of a plurality of pieces of second data different from the plurality of pieces of first data. (“the goal is to train a classifier for each of these target domains. As we do not have labelled data from the target domains, we will need to perform domain adaptation.” Column 14, lines 48-50; we perform a domain adaptation step between T1 and T2, wherein we adapt the weights of M.sub.T1 to output a model for T2, namely M.sub.T2.” Column 15, lines 11-13 (Gan)) (Since the domains differ, the labelling would also differ.)
Regarding claim 6, Gan, as modified by Liang, teaches the learning system according to claim 5, wherein the at least one processor is configured to: create a third learning model capable of the labeling based on a third group being a group of pieces of the second data labeled by the second learning model; (“we perform a domain adaptation step between T1 and T2, wherein we adapt the weights of M.sub.T1 to output a model for T2, namely M.sub.T2.” Column 15, lines 11-13; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The process that creates the first and second model can be repeated to create a third model.) convert a fourth group being a group of pieces of the second data which are not labeled by the second learning model so that a distribution of the fourth group is close to a distribution of the third group; (“the control module 96 uses a selection technique based on computing domain divergence, i.e. how far are two domains from each other. In particular, the control module 96 may determine a pairwise divergence (T.sub.2, S) or (T.sub.2, T.sub.1) between the target domain (i e T2) and the candidate domains (S and T1), and identify the pair with the smallest divergence.” Column 14, lines 64-67, and Column 15, lines 1-3; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The steps can be repeated to convert pieces of data that do not fit in the first, second, or third domain into a fourth group.) and execute the labeling for the fourth group based on the third learning model and the fourth group (“we perform a domain adaptation step between T1 and T2, wherein we adapt the weights of M.sub.T1 to output a model for T2,” Column 15, lines 11-13; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (With the repeatable steps, a new model is made to label the fourth group.)
Regarding claim 7, Gan, as modified by Liang, teaches the learning system according to claim 4, wherein the at least one processor is configured to create, based on the first group, the third group, and the fourth group a fourth learning model which is different from any one of the first learning model, the second learning model, or the third learning model and which is capable of the labeling. (“we perform a domain adaptation step between T1 and T2, wherein we adapt the weights of M.sub.T1 to output a model for T2,” Column 15, lines 11-13; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (With the repeatable steps, a fourth model is made to label the fourth group.)
Regarding claim 8, Gan, as modified by Liang, teaches the learning system according to claim 7, wherein the at least one processor is configured to determine, based on similarity between the distribution of the first group and the distribution of the third group, whether to use the first group to create the fourth learning model, create the fourth learning model without based on the first group when to use the first group, and to create the fourth learning model based on the first group when it is determined to use the first group. (“the control module 96 uses a selection technique based on computing domain divergence, i.e. how far are two domains from each other. In particular, the control module 96 may determine a pairwise divergence (T.sub.2, S) or (T.sub.2, T.sub.1) between the target domain (i e T2) and the candidate domains (S and T1), and identify the pair with the smallest divergence.” Column 14, lines 64-67, and Column 15, lines 1-3; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The steps can be repeated to determine whether pieces of data do or do not fit in the first, second, third, fourth, etc. domain, with a domain’s creation (and model’s creation) depending on the smallest divergence between the domains, which could include a divergence between a first and third group, not a first group, or a first group and source data.)
Regarding claim 9, Gan, as modified by Liang, teaches the learning system according to claim 7, wherein the at least one processor is configured to create the fourth learning model further based on the second group (“the control module 96 uses a selection technique based on computing domain divergence, i.e. how far are two domains from each other. In particular, the control module 96 may determine a pairwise divergence (T.sub.2, S) or (T.sub.2, T.sub.1) between the target domain (i e T2) and the candidate domains (S and T1), and identify the pair with the smallest divergence.” Column 14, lines 64-67, and Column 15, lines 1-3; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The steps can be repeated to create a fourth model based on a divergence which uses a second group of data.)
Regarding claim 10, Gan, as modified by Liang, teaches the learning system according to claim 9, wherein the at least one processor is configured to determine, based on similarity between the distribution of the second group and the distribution of the fourth group, whether to use the second group to create the fourth learning model, create the fourth learning model without based on the second group when to use the second group, and to create the fourth learning model based on the second group when to use the second group. (“the control module 96 uses a selection technique based on computing domain divergence, i.e. how far are two domains from each other. In particular, the control module 96 may determine a pairwise divergence (T.sub.2, S) or (T.sub.2, T.sub.1) between the target domain (i e T2) and the candidate domains (S and T1), and identify the pair with the smallest divergence.” Column 14, lines 64-67, and Column 15, lines 1-3; “Now if T3 is introduced in the system, the steps above can be repeated in order to select the appropriate candidate encoder for the new domain and eventually train a model M.sub.T3” Column 15, lines 12-16 (Gan)) (The steps can be repeated to determine whether pieces of data do or do not fit in the first, second, third, fourth, etc. domain, with a domain’s creation (and model’s creation) depending on the smallest divergence between the domains, which could include a divergence between a second and fourth group, not a second group, or a second group and source data.)
Regarding claim 11, Gan, as modified by Liang, teaches the learning system according to claim 1, wherein the at least one processor, is configured to create the second learning model based on the second group and before being converted. (“The means for selecting the initial instance of the source encoder may select an optimum source encoder from the plurality of candidate encoders according to some metric. The said metric may be related to a divergence (e.g. a computing domain divergence) between domains. Thus, the source encoder may be selected based on the candidate encoder that shows the smallest divergence between domains, according to the chosen metric (i.e. the smallest difference between the domain of the candidate and the domain of the source).” Column 2, lines 33-42 (Gan))
Regarding claim 12, Gan, as modified by Liang, teaches the learning system according to claim 1, wherein the at least one processor is configured to execute, based on the second group, additional learning for the first learning model which has learned the first group. (“ADDA adopts an adversarial training approach which involves training a discriminator D and a feature encoder M.sub.T using adversarial training. The goal of the discriminator is to distinguish the two domains while the goal of the encoder is to fool the discriminator by generating features that are indistinguishable across domains. Both the encoder and discriminator participate in this game, trying to fool each other, and in the process, both get better at their respective tasks. More importantly, our encoder M.sub.T may learn to map target domain data into the same feature space as the source encoder, thereby enhancing the inference accuracy in the target domain.”, Column 15, lines 56-67 (Gan))
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Gan in view of Liang, and in further view of Wu et al. (Herein referred to as Wu) (Confusing the Crowd: Task Instruction Quality on Amazon Mechanical Turk)
Regarding claim 13, Gan, as modified by Liang, teaches the learning system according to claim 1 but does not explicitly teach to provide the pieces of the first data satisfying the first condition to an administrator who executes the labeling; receive specification for the label by the administrator, nor execute the labeling for the first group based on the specification by the administrator.
Wu teaches to provide the pieces of the first data satisfying the first condition to an administrator who executes the labeling; receive specification for the label by the administrator, and execute the labeling for the first group based on the specification by the administrator. (“The questions used to elicit workers’ assessments of tasks (Figure 1) were based on two types of metrics: descriptive (properties of the task itself) and prospective (workers’ prediction of outcomes related to accuracy, worker acceptability, etc.).”, pg. 3, under Instruction Metrics; See also Figure 1 and Prospective metrics) (A worker, who acts as an administrator is given the task to label a group of data based on specification and the provided data.)
Therefore, it would have been considered obvious to one of ordinary skill in the art, prior to the current application’s filing date, to implement an administrator to label data based on provided data, as described by Wu, with the provided data and labels of Gan, as modified by Liang. One would be motivated to combine the teachings, prior to the filing date of the current application, as Wu’s task design can lead to better outcomes for workers and quality metrics, as described by Wu. (“The key contributions of this work are as follows: 1. Evaluations of tasks scraped from Mechanical Turk measured (a) workers’ assessment of the task quality, and (b)adherence to established best practices for task design. 2. Actual effects of specific best practices on desired out-comes (e.g., accuracy, worker acceptability, etc.) were measured for a single task by creating systematic mutations and measuring workers perceptions (by inspection)and actual performance (by posting the tasks). 3. Results of the experiments show that (a) adherence to best practices can affect outcome, but that (b) workers are more resilient to flaws in task quality than current popular belief might suggest.”, pg. 1, right column)
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Gan in view of Liang and in further view of SARA MAKKI et al. (Herein referred to as Makki) (An Experimental Study With Imbalanced Classification Approaches for Credit Card Fraud Detection)
Regarding claim 14, Gan, as modified by Liang, teaches the learning system according to claim 1 but does not explicitly teach each of the plurality of pieces of first data indicates an action of a user who uses a predetermined service, wherein the predetermined service is provided based on user information on the user, wherein the labeling is processing of determining whether the action of the user having valid user information is fraudulent, nor wherein the label is a determined fraud label indicating that the fraud is determined.
Makki teaches each of the plurality of pieces of first data indicates an action of a user who uses a predetermined service, wherein the predetermined service is provided based on user information on the user, (“The dataset used in our experiment contains credit card fraud labeled data”, pg. 6, left column, under Dataset and Variable Selection) (The data pertains to a user who uses credit card services which contains user information, such as ID and gender.) wherein the labeling is processing of determining whether the action of the user having valid user information is fraudulent (“Statistical characteristics of the numerical variables are shown in Table 1. In the data set, gender’s frequencies are, respectively, 6,178,231 male (61.7%) and 3,821,769 female (38.3%). Moreover, 596,014 (5.96%) are fraud cases and 9,403,986 (94.04%) are legitimate.” Pg. 7, left column, first paragraph) wherein the label is a determined fraud label indicating that the fraud is determined. (“This approach uses only one class of the data (usually the minority class) and learns its characteristics. In our case, the classification takes place in the testing phase. After training the algorithm with one class, it should be able to determine whether a certain transaction belongs to the minority class or not.”, pg. 5, right column, under One-Class Classification (OCC))
Therefore, it would have been considered obvious to one of ordinary skill in the art, prior to the current application’s filing date, to combine the method of Gan, as modified by Liang, with the fraud detection of Makki. One would have been motivated to combine the teachings, prior to the effective filing date, as Makki’s findings help identify the weaknesses of fraud labeled datasets, to help prevent false alarms, or inaccurate labeling. (“We explored these solutions along with the machine learning algorithms used for fraud detection. We identified their weaknesses and summarized the results that we obtained using a credit card fraud labeled dataset. According to this paper, imbalanced classification approaches are ineffective, especially when the data are highly imbalanced. This paper reveals that the existing approaches result in a large number of false alarms, which are costly to financial institutions. This may lead to inaccurate detection as well as increasing the occurrence of fraud cases.”, pg. 1, Abstract)
Response to Arguments
Applicant's arguments filed on February 27th, 2026 have been fully considered but they are not persuasive. The applicant argues in substance:
Argument 1: The claims are allowable under 101 as it solves a technical problem with a technical solution.
The examiner respectfully disagrees. While the applicant ‘s problem is technical in nature, the claim language, under the broadest reasonable interpretation, points to an abstract idea, and not a technical solution, for reasons explained above.
Argument 2: The specific create/training of the machine learning model cannot be performed mentally and is therefore patent eligible subject matter.
The examiner respectfully disagrees. Under MPEP 2106.04(a)(2), claims can recite a mental process even if they are claimed as being performed on a computer. The training/creating of a machine learning model is interpreted by the examiner to be, under the broadest reasonable interpretation, a mental process able to be performed on a generic computer, in a computing environment, or with a computer as a tool to perform the mental process. As such, the training/creating step is interpreted to be an abstract idea.
Argument 3: The amended limitation make the invention allowable, as it further clarifies an improvement to the computer system, such as the adjusting of parameters of the second learning model, which cannot be performed mentally and reflects a technical improvement.
The examiner respectfully disagrees. For reasons similar to argument 2, the “adjusting of parameters of the second learning model” is interpreted to be, under the broadest reasonable interpretation, a mental process able to be performed on a generic computer, or in a computing environment, or with a computer as a tool to perform the mental process. It does not reflect an improvement in a technical field, but rather reflects an abstract idea, with the additional features and limitations merely amounting to an improvement of an abstract idea. Therefore, the arguments are not persuasive and the 101 rejections are maintained.
Argument 4: Claim 1 as amended is not anticipated or rendered obvious by cited references.
Applicant’s arguments with respect to claim 1 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Claims 15 and 16 are rejected under a similar rationale. Therefore the 103 rejections are maintained.
Conclusion
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 Tyler E Iles whose telephone number is (571)272-5442. The examiner can normally be reached 9:00am - 5:00pm.
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, Kakali Chaki can be reached at (571) 272-3719. 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.
/T.E.I./ Patent Examiner, Art Unit 2122
/KAKALI CHAKI/ Supervisory Patent Examiner, Art Unit 2122
Prosecution Insights