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 .
Priority
Acknowledgement is made of Applicant’s claim of priority from EP23203859, filed October 16, 2023.
Information Disclosure Statement
The information disclosure statements (“IDS”) filed on October 14, 2024 was reviewed and the listed references were noted.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-7, 10-12 and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over Yuliya Rudenko (US 2015/0177842 A1) in view of Namsoon Jung (US 2021/0056412 A1) further in view of Lyons et al. (US 2023/0316136 A1) and Narayanam et al. (US 2024/0320538 A1, filed March 20, 2023).
Regarding claim 1, Rudenko teaches a method of operating an
obtaining at least one feature vector that encodes measurement data provided by a depth sensor for a gesture executed by a user (Rudenko, Para. [0033], a sensor (such as a HD 3D depth sensor or related device. Para. [0035], the authentication system 130 processes the depth images (depth maps) and retrieves 3D gesture data, which may include a series of 3D coordinates associated with a virtual skeleton joints or a series of 3D coordinates associated with finger cushions, or similar/related information. The 3D gesture data are then processed to generate a feature vector (i.e., feature vector that encodes measurement data provided by a depth sensor)).
using a machine-learning model, inferring, from the at least one feature vector, a gesture class prediction associated with the gesture (Rudenko, Para. [0034], the present technology may acquire 3D user-gesture data (i.e., feature vector), analyze it using one or more machine learning algorithms (i.e., machine-learning model) and determine a gesture type (i.e., gesture class prediction)).
Although Rudenko teaches a computing device having processors to be operatively coupled to or embed the depth sensing sensor(s) and/or video camera(s) (Rudenko, Para. [0044]), Rudenko does not explicitly teach that the processor is “an edge-deployed processor”. However, in an analogous field of endeavor, Jung teaches the one or more processors corresponds to one or more neuromorphic processors (NMP) that are deployed on the edge device (Jung, Para. [0036]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Rudenko with the teachings of Jung by including operating an edge-deployed processor. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for edge device to support neural network transfer learning capability, as recognized by Jung.
Although Rudenko in view of Jung teaches determining a gesture type from a feature vector (Rudenko, Para. [0034]), they do not explicitly teach “determining at least one feature relevance vector for the at least one feature vector, each of the at least one feature relevance vector comprising feature relevance values, and each of the feature relevance values being indicative of a dependency of the gesture class prediction on respective one or more feature values of the at least one feature vector”. However, in an analogous field of endeavor, Lyons teaches that a set of Shapley values (i.e., feature relevance values) can be generated for each input feature (e.g., feature vector) where the set of Shapley values =
[
s
h
a
p
1
,
s
h
a
p
2
,
…
,
s
h
a
p
n
]
(i.e., feature relevance vector). More specifically, there can be n input features, and
s
h
a
p
i
is defined for input feature i. For each
s
h
a
p
i
, it is determined whether the input feature i satisfies a threshold condition, and the input feature i is selected for the subset of features if the
s
h
a
p
i
i satisfies the threshold condition (Lyons, Para. [0042]). The subset of features can include one or more features determined to be sufficiently important for classifying the activity (Lyons, Para. [0040]) (i.e., feature relevance value is indicative of dependency of class on one or more feature values in the feature vector).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date to modify the method of Rudenko in view of Jung with the teachings of Lyons by including determine a feature relevance vector (i.e., a vector of Shapley values) that indicates the important features from the feature vector for classifying an activity (i.e., is indicative of dependency of the gesture classification of Rudenko on the one or more feature values in the feature vector). One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for generating a machine learning model to recognize activities being performed by a human, as recognized by Lyons.
Although Rudenko in view of Jung further in view of Lyons teaches determining a gesture type from a feature vector (Rudenko, Para. [0034]), they do not explicitly teach “determining a user output associated with the gesture based on the at least one feature relevance vector” and “controlling a user interface to provide the user output to the user”. However, in an analogous field of endeavor, Narayanam teaches a user interface could be provided to the user device that lists anomalous data subsets identifying the subset of attributes and records in each anomalous data subset (i.e., user output associated with the gesture) or that provides a view of the tabular data in which the tabular data has been restructured (e.g., with the most anomalous attributes and records moved towards the top left) with a visual indicator applied to each anomalous data subset (i.e., controls a user interface to provide a user output associated with the gesture) (Narayanam, Para. [0063]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Rudenko in view of Jung further in view of Lyons with the teachings of Narayanam by including providing a user interface for a user output that identifies anomalous data associated with the gesture. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for a computing device for gesture detection and recognition, as recognized by Narayanam. Thus, the claimed invention would have been obvious to one having ordinary skill in the art before the effective filing date.
Regarding claim 2, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 1, further comprising classifying the gesture as anomalous or non-anomalous, wherein the user output is selectively provided to the user responsive to classifying the gesture as anomalous (Narayanam, Para. [0032], anomaly scores that reflect an extent to which each attribute and record includes anomalous data (i.e., classifies the gesture as anomalous or non-anomalous). In some aspects, the anomaly scores comprise Shapley values computed for the attributes and records. Para. [0069], output identifying the anomalous data subsets is provided. In some configurations, the output comprises an indication of attributes and records for each anomalous data subset, and the anomalous data subsets can be ordered based on the extent of their anomalous data).
The proposed combination as well as the motivation for combining the Rudenko, Jung, Lyons and Narayanam references presented in the rejection of Claim 1, apply to Claim 2 and are incorporated herein by reference. Thus, the method recited in Claim 2 is met by Rudenko in view of Jung further in view of Lyons and Narayanam.
Regarding claim 3, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 2, wherein the classifying of the gesture as anomalous or non-anomalous is based on the at least one feature relevance vector (Narayanam, Para. [0032], the anomaly scores comprise Shapley values computed for the attributes and records).
The proposed combination as well as the motivation for combining the Rudenko, Jung, Lyons and Narayanam references presented in the rejection of Claim 1, apply to Claim 3 and are incorporated herein by reference. Thus, the method recited in Claim 3 is met by Rudenko in view of Jung further in view of Lyons and Narayanam.
Regarding claim 4, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 2, wherein the classifying of the gesture as anomalous or non-anomalous is based on an uncertainty measure associated with the inferring of the gesture class prediction using the machine-learning model (Narayanam, Para. [0018], an anomaly score for a data element is a value indicative of a likelihood that the data element is anomalous (i.e., uncertainty measure of prediction). The anomaly score for a data element, for instance, can be based on a prediction or reconstruction error determined using a machine learning model (e.g., an autoencoder) trained to predict values of attributes for records given values for other attributes).
The proposed combination as well as the motivation for combining the Rudenko, Jung, Lyons and Narayanam references presented in the rejection of Claim 1, apply to Claim 4 and are incorporated herein by reference. Thus, the method recited in Claim 4 is met by Rudenko in view of Jung further in view of Lyons and Narayanam.
Regarding claim 5, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 2, wherein the classifying of the gesture as anomalous or non-anomalous is based on a distance of the at least one feature vector to one or more predefined reference feature vectors (Rudenko, Para. [0035], machine learning algorithms enable determining similarity values between a first feature vector and each of the plurality of second, reference feature vectors (which similarity value, or representation, can be as simple as a difference vector between a first vector and a second vector) (i.e., a distance of the at least one feature vector to one or more predefined reference feature vectors). If the similarity value (also referred herein as to a score or rank) is above a predetermined threshold, then the authentication system determines that the feature vector relates to a pre-validated user that is associated with the most similar feature vector. Thus, the user is authenticated. Otherwise, the user is not authenticated (i.e., classifying the gesture as anomalous or non-anomalous)).
Regarding claim 6, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 2, wherein the classifying of the gesture as anomalous or non-anomalous is based on a threshold comparison between each of the feature values of the at least one feature vector and respective thresholds (Narayanam, Para. [0037], a binary labeling approach is used in which each data element is labeled with either a first label indicating the data element is likely anomalous (e.g., labeled “anomalous” or “potential anomaly”) or a second label indicating the data element is not likely anomalous (e.g., labeled “not anomalous” or “not a potential anomaly”). In some configurations, labels are assigned based on comparison of anomaly scores for data elements to a threshold. For instance, one label is assigned to a data element if the anomaly score satisfies the threshold, while the other label is assigned to the data element if the anomaly score does not satisfy the threshold).
The proposed combination as well as the motivation for combining the Rudenko, Jung, Lyons and Narayanam references presented in the rejection of Claim 1, apply to Claim 6 and are incorporated herein by reference. Thus, the method recited in Claim 6 is met by Rudenko in view of Jung further in view of Lyons and Narayanam.
Regarding claim 7, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 1, wherein the user output is provided to the user as part of a re-training process for populating a training dataset for re-training the machine-learning model (Lyons, Para. [0056], generating the subset of features can include determining, for each feature, whether the respective Shapley value satisfies a threshold condition (e.g., is greater than or equal to a threshold value). If so, the feature can be added to the subset of features. If not, the feature is not included in the subset of features. That is, the feature is not used to implement the machine learning model (e.g., train the machine learning model)).
The proposed combination as well as the motivation for combining the Rudenko, Jung, Lyons and Narayanam references presented in the rejection of Claim 1, apply to Claim 7 and are incorporated herein by reference. Thus, the method recited in Claim 7 is met by Rudenko in view of Jung further in view of Lyons and Narayanam.
Regarding claim 10, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 1, wherein the user output is indicative of one or more feature values of the at least one feature vector that are associated with feature relevance values that deviate from a predefined reference or exceed or fall below a predefined threshold (Rudenko, Para. [0035], machine learning algorithms enable determining similarity values between a first feature vector and each of the plurality of second, reference feature vectors (which similarity value, or representation, can be as simple as a difference vector between a first vector and a second vector) (i.e., a distance of the at least one feature vector to one or more predefined reference feature vectors). If the similarity value (also referred herein as to a score or rank) is above a predetermined threshold, then the authentication system determines that the feature vector relates to a pre-validated user that is associated with the most similar feature vector. Thus, the user is authenticated. Otherwise, the user is not authenticated (i.e., classifying the gesture as anomalous or non-anomalous)).
Regarding claim 11, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 10, wherein the predefined reference or the predefined threshold are determined based on previously determined feature relevance vectors (Rudenko, Para. [0035], machine learning algorithms enable determining similarity values between a first feature vector and each of the plurality of second, reference feature vectors (which similarity value, or representation, can be as simple as a difference vector between a first vector and a second vector) (i.e., a distance of the at least one feature vector to one or more predefined reference feature vectors). If the similarity value (also referred herein as to a score or rank) is above a predetermined threshold, then the authentication system determines that the feature vector relates to a pre-validated user that is associated with the most similar feature vector. Thus, the user is authenticated. Otherwise, the user is not authenticated (i.e., classifying the gesture as anomalous or non-anomalous)).
Regarding claim 12, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 1, wherein the at least one feature vector comprises feature values in one or more of the following dimensions: range of a gesture object; velocity of the gesture object; angular orientation of the gesture object; azimuthal angle of the gesture object; or elevation angle of the gesture object (Rudenko, Para. [0050], every snapshot or image may be pre-processed to retrieve one or more features associated with the 3D user hand gesture. The features of a single snapshot may be associated with one or more of the following: a hand posture, a hand shape, fingers' postures, fingers' positions (i.e., 3D coordinates), finger cushions' postures, finger cushions' positions (i.e., 3D coordinates), angles between fingers, rotational angles of hand palm, a velocity of motion of one or more fingers or hand, acceleration of motion of one or more fingers or hand, dimensions of fingers, lengths between various finger cushions, and/or other aspects or manners of hand and/or finger configuration and/or movement).
Claims 16-20 recite systems with elements corresponding to the steps recited in Claims 1-7, respectively. Therefore, the recited elements of these claims are mapped to the proposed combination in the same manner as the corresponding steps in their corresponding method claims. Additionally, the rationale and motivation to combine the Rudenko, Jung, Lyons and Narayanam references, presented in rejection of Claim 1, apply to these claims. Finally, the combination of the Rudenko, Jung, Lyons and Narayanam references discloses a processor and a memory (Rudenko, Para. [0065], the example computer system 1500 includes a processor or multiple processors 1505 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), or both), and a main memory 1510 and a static memory 1515, which communicate with each other via a bus 1520).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Yuliya Rudenko (US 2015/0177842 A1) in view of Namsoon Jung (US 2021/0056412 A1) further in view of Lyons et al. (US 2023/0316136 A1) and Narayanam et al. (US 2024/0320538 A1, filed March 20, 2023), as applied to claims 1-7, 10-12 and 16-20 above, and further in view of Sharpe et al. (US 2025/0053491 A1, filed August 10, 2023).
Regarding claim 8, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 1, further comprising:
re-training the machine-learning model based on the training dataset, to thereby obtain the machine-learning model (Lyons, Para. [0056], generating the subset of features can include determining, for each feature, whether the respective Shapley value satisfies a threshold condition (e.g., is greater than or equal to a threshold value). If so, the feature can be added to the subset of features. If not, the feature is not included in the subset of features. That is, the feature is not used to implement the machine learning model (e.g., train the machine learning model)).
The proposed combination as well as the motivation for combining the Rudenko, Jung, Lyons and Narayanam references presented in the rejection of Claim 1, apply to Claim 7 and are incorporated herein by reference.
Although Rudenko in view of Jung further in view of Lyons and Narayanam teaches retraining based on the output (Lyons, Para. [0056]), they do not explicitly teach “controlling the user interface to obtain a user input associated with the user output” and “based on the user input, selectively including the at least one feature vector in a training dataset”. However, in an analogous field of endeavor, Sharpe teaches outputs may be fed back to model as input to train model (e.g., alone or in conjunction with user indications of the accuracy of outputs, labels associated with the inputs, or with other reference feedback information (i.e., user input associated with the user output)). For example, the system may receive a first labeled feature input, wherein the first labeled feature input is labeled with a known prediction for the first labeled feature input. The system may then train the first machine learning model to classify the first labeled feature input with the known prediction (e.g., predicting resource allocation values for user systems) (Sharpe, Para. [0034]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date to modify the method of Rudenko in view of Jung further in view of Lyons and Narayanam with the teachings of Sharpe by including controlling the user interface to obtain a user input associated with the output and including the feature input to train the machine learning model. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for improving accuracy and benefitting further predictions, as recognized by Sharpe. Thus, the claimed invention would have been obvious to one having ordinary skill in the art before the effective filing date.
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Yuliya Rudenko (US 2015/0177842 A1) in view of Namsoon Jung (US 2021/0056412 A1) further in view of Lyons et al. (US 2023/0316136 A1), Narayanam et al. (US 2024/0320538 A1, filed March 20, 2023) and Sharpe et al. (US 2025/0053491 A1, filed August 10, 2023) as applied to claim 8 above, and further in view of Yoo et al. (US 2021/0273707 A1).
Regarding claim 9, Rudenko in view of Jung further in view of Lyons, Narayanam and Sharpe teaches the method of claim 8, as described above.
Although Rudenko in view of Jung further in view of Lyons, Narayanam and Sharpe teaches retraining the model (Lyons, Para. [0056]), they do not explicitly teach “providing an uplink message to a central server, the uplink message being indicative of weights of the machine-learning model upon completing the re-training”. However, in an analogous field of endeavor, Yoo teaches an uplink channel that refers to the communication link from the user equipment to the base station (Yoo, Para. [0003]). The uplink channel transmits one or more encoder structures of the neural network model and the one or more encoder weights (Yoo, Para. [0108]).
Therefore, it would have been obvious to one having ordinary skill in the art to modify the method of Rudenko in view of Jung further in view of Lyons, Narayanam and Sharpe with the teachings of Yoo by including providing an uplink message to a central server of the encoder weights. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for channel state information feedback, as recognized by Yoo. Thus, the claimed invention would have been obvious to one having ordinary skill in the art before the effective filing date.
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over Yuliya Rudenko (US 2015/0177842 A1) in view of Namsoon Jung (US 2021/0056412 A1) further in view of Lyons et al. (US 2023/0316136 A1) and Narayanam et al. (US 2024/0320538 A1, filed March 20, 2023), as applied to claims 1-7, 10-12 and 16-20 above, and further in view of Yoo et al. (US 2021/0273707 A1).
Regarding claim 13, Rudenko in view of Jung further in view of Lyons and Narayanam teaches the method of claim 1, as described above.
Although Rudenko in view of Jung further in view of Lyons and Narayanam teaches retraining the model (Lyons, Para. [0056]), they do not explicitly teach “obtaining, from a central server, a downlink message indicative of weights of the machine-learning model prior to the inferring”. However, in an analogous field of endeavor, Yoo teaches a downlink channel for transmitting one or more decoder weights of the neural network model (Yoo, Para. [0107]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Rudenko in view of Jung further in view of Lyons and Narayanam with the teachings of Yoo by including a downlink channel to send a message indicative of weights of the machine-learning model. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for channel state information feedback, as recognized by Yoo. Thus, the claimed invention would have been obvious to one having ordinary skill in the art before the effective filing date.
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. (US 2021/0110261 A1) in view of Yuliya Rudenko (US 2015/0177842 A1) further in view of Namsoon Jung (US 2021/0056412 A1) and Lior Aronovich (US 2024/0333821 A1, filed March 27, 2023).
Regarding claim 14, Lee teaches a method of operating a central server, the method comprising:
obtaining, (Lee, Para. [0115], the weight of the neural network may be transmitted to the base station via an uplink channel),
providing, (Lee, Para. [0119], downlink channel feedback of the weights of which are updated).
Although Lee teaches updating weights of a neural network model (Lee, Para. [0119]), Lee does not explicitly teach “the machine-learning model being used, (Rudenko, Para. [0034]). Rudenko further teaches a sensor (such as a HD 3D depth sensor or related device (Rudenko, Para. [0033]). The authentication system 130 processes the depth images (depth maps) and retrieves 3D gesture data, which may include a series of 3D coordinates associated with a virtual skeleton joints or a series of 3D coordinates associated with finger cushions, or similar/related information. The 3D gesture data are then processed to generate a feature vector (Rudenko, Para. [0035]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date to modify the method of Lee with the teachings of Rudenko by including a machine-learning model for inferring gesture class predictions from feature vectors representing measurement data from depth sensors. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for recognizing 3D hand gestures, as recognized by Rudenko.
Although Lee in view of Rudenko teaches a processor of the BS and the UE (Lee, Para. [0160]), they do not explicitly teach “edge-deployed processors”. However, in an analogous field of endeavor, Jung teaches the one or more processors corresponds to one or more neuromorphic processors (NMP) that are deployed on the edge device (Jung, Para. [0036]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Lee in view of Rudenko with the teachings of Jung by including operating an edge-deployed processor. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for edge device to support neural network transfer learning capability, as recognized by Jung.
Although Lee in view of Rudenko further in view of Jung teaches (Lee, Para. [0119]), they do not explicitly teach “consolidating the weights, to determine updated weights of the machine-learning model”. However, in an analogous field of endeavor, Aronovich teaches federated learning is an approach in which each device (e.g., edge computing) receives the current model and computes an updated model locally using its local data. The locally trained models are then sent from the devices back to the central server where they are aggregated (or globally trained), for example, by averaging weights, and then a single consolidated and improved global model is sent back to the devices (Aronovich, Para. [0024]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date to modify the method of Lee in view of Rudenko further in view of Jung with the teachings of Aronovich by including that updating the weights includes consolidating the weights to generate an improved global model. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for reducing bandwidth requirements for data transmissions, as recognized by Aronovich. Thus, the claimed invention would have been obvious to one having ordinary skill in the art before the effective filing date.
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. (US 2021/0110261 A1) in view of Yuliya Rudenko (US 2015/0177842 A1) further in view of Namsoon Jung (US 2021/0056412 A1) and Lior Aronovich (US 2024/0333821 A1, filed March 27, 2023), as applied to claim 14 above, and further in view of Lyons et al. (US 2023/0316136 A1) and Narayanam et al. (US 2024/0320538 A1, filed March 20, 2023).
Regarding claim 15, Lee in view of Rudenko further in view of Jung and Aronovich teaches the method of claim 14, wherein each of the multiple edge-deployed processors is configured to:
obtain at least one feature vector that encodes measurement data provided by a depth sensor for a gesture executed by a user (Rudenko, Para. [0033], a sensor (such as a HD 3D depth sensor or related device. Para. [0035], the authentication system 130 processes the depth images (depth maps) and retrieves 3D gesture data, which may include a series of 3D coordinates associated with a virtual skeleton joints or a series of 3D coordinates associated with finger cushions, or similar/related information. The 3D gesture data are then processed to generate a feature vector (i.e., feature vector that encodes measurement data provided by a depth sensor)),
use the machine-learning model to infer from the at least one feature vector, a gesture class prediction associated with the gesture (Rudenko, Para. [0034], the present technology may acquire 3D user-gesture data (i.e., feature vector), analyze it using one or more machine learning algorithms (i.e., machine-learning model) and determine a gesture type (i.e., gesture class prediction)).
The proposed combination as well as the motivation for combining the Lee, Rudenko, Jung and Aronovich references presented in the rejection of Claim 14, apply to Claim 15 and are incorporated herein by reference.
Although Lee in view of Rudenko further in view of Jung and Aronovich teaches determining a gesture type from a feature vector (Rudenko, Para. [0034]), they do not explicitly teach “determine at least one feature relevance vector for the at least one feature vector, each of the at least one feature relevance vector comprising feature relevance values, and each of the feature relevance values being indicative of a dependency of the gesture class prediction on respective one or more feature values of the at least one feature vector”. However, in an analogous field of endeavor, Lyons teaches that a set of Shapley values (i.e., feature relevance values) can be generated for each input feature (e.g., feature vector) where the set of Shapley values =
[
s
h
a
p
1
,
s
h
a
p
2
,
…
,
s
h
a
p
n
]
(i.e., feature relevance vector). More specifically, there can be n input features, and
s
h
a
p
i
is defined for input feature i. For each
s
h
a
p
i
, it is determined whether the input feature i satisfies a threshold condition, and the input feature i is selected for the subset of features if the
s
h
a
p
i
i satisfies the threshold condition (Lyons, Para. [0042]). The subset of features can include one or more features determined to be sufficiently important for classifying the activity (Lyons, Para. [0040]) (i.e., feature relevance value is indicative of dependency of class on one or more feature values in the feature vector).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date to modify the method of Lee in view of Rudenko further in view of Jung and Aronovich with the teachings of Lyons by including determine a feature relevance vector (i.e., a vector of Shapley values) that indicates the important features from the feature vector for classifying an activity (i.e., is indicative of dependency of the gesture classification of Rudenko on the one or more feature values in the feature vector). One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for generating a machine learning model to recognize activities being performed by a human, as recognized by Lyons.
Although Lee in view of Rudenko further in view of Jung, Aronovich and Lyons teaches determining a gesture type from a feature vector (Rudenko, Para. [0034]), they do not explicitly teach “determine a user output associated with the gesture based on the at least one feature relevance vector” and “control a user interface to provide the user output to the user”. However, in an analogous field of endeavor, Narayanam teaches a user interface could be provided to the user device that lists anomalous data subsets identifying the subset of attributes and records in each anomalous data subset (i.e., user output associated with the gesture) or that provides a view of the tabular data in which the tabular data has been restructured (e.g., with the most anomalous attributes and records moved towards the top left) with a visual indicator applied to each anomalous data subset (i.e., controls a user interface to provide a user output associated with the gesture) (Narayanam, Para. [0063]).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Lee in view of Rudenko further in view of Jung, Aronovich and Lyons with the teachings of Narayanam by including providing a user interface for a user output that identifies anomalous data associated with the gesture. One having ordinary skill in the art would have been motivated to combine these references because doing so would allow for a computing device for gesture detection and recognition, as recognized by Narayanam. Thus, the claimed invention would have been obvious to one having ordinary skill in the art before the effective filing date.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Emma Rose Goebel whose telephone number is (703)756-5582. The examiner can normally be reached Monday - Friday 7:30-5.
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, Amandeep Saini can be reached at (571) 272-3382. 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.
/Emma Rose Goebel/Examiner, Art Unit 2662
/AMANDEEP SAINI/Supervisory Patent Examiner, Art Unit 2662
Prosecution Insights