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 .
Claim Objections
Claim(s) 1, 11, and 17 objected to because of the following informalities:
In claim 1, “on input data” should read “the input data”.
In claim 11, “the neural network” should read “a neural network”
In claim 17, “connected first” should read “connected to the first”.
In claim 17, “the video feed the second AI system” should read “the video feed or the second AI system”
Appropriate correction is required.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: “configured to” first recited in claim 1, line 5.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim(s) 1, 6, 11, and 17 rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claim(s) 1, 6, 11, and 17 recite the limitation “configured to” perform functions (e.g., store and manage data, detect and receive environmental feedback, perform a training improvement method, etc.). The specification does not disclose sufficient corresponding structure for the recited “configured to” limitation(s). For example, paragraph [0023] describes the first, second and third AI systems capable of receiving data, determining events, processing input and predicting outcomes, but merely restates the claimed functions without providing any specific algorithm, or defined structure for performing these functions.
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.
Claim(s) 1-17 is/are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
In regards to claim(s) 1, the claim recites multiple limitations using functional language (e.g., “configured to… receive the environmental feedback,” “process the input data,” “analyze the input data,” “use captured data to predict expected outcomes,” “autonomously provide coaching advice”), which renders the claim indefinite. It is unclear what constitutes how the functions are performed, as these functions may be implemented using different techniques.
Claim(s) 2-17 are rejected for their dependency on claim 1.
In regards to claim(s) 4, the claim recites “including, but not limited to,” which renders the claim indefinite. The phrase “data captured from sources including, but not limited to, video feeds from cameras, audio data, wearable fitness trackers, smartwatch data, and wearable biomechanical sensors.,” introduces an open-ended and undefined set of data sources.
In regards to claim(s) 6, the claim recites “the first AI system is… configured to assign multiple event labels simultaneously depending on the complexity of the event data,”. Similar to above to above, the term “configured to” renders the claim indefinite. The claim does not specify how these functions are achieved.
Claim(s) 11 and 17 have the same limitations as claim 6 and are rejected for the same reasons.
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) 1-17 are rejected under 35 U.S.C. 101 because the claimed invention us directed to an abstract idea without significantly more
[STEP 1] The claim recites at least one step or structure. Thus, the claim is to a machine and process which is one of the statutory categories of invention.
[STEP2A PRONG 1] The claims recite limitations which have been construed as certain methods of organizing human activity and/or mental processes. Claim(s) 1 recite a system, claim(s) 11 and 17 further recite method steps performed by the system, for performing steps including receiving data associated with a target player, processing the data using one or more artificial intelligence systems, determining characteristics and/or performance attributes of the target player, predicting outcomes or consequences of actions, and generating training improvement recommendations or assessments. Under their broadest reasonable interpretation, the recited steps establish collecting information (e.g., video data and physical feedback), evaluating that information (e.g., analyzing performance, predicting outcomes, and determining training categories), and providing results (e.g., recommendations, assessments). The recited steps further include a mental process. Evaluating an individual’s performance, predicting outcomes, and providing feedback or advice are longstanding human activities that can be performed mentally or with pen and paper (see MPEP 2106.04(a)(2)). Accordingly, claims 1-17 recite abstract ideas.
[STEP2A PRONG II] The judicial exception is not integrated into a practical application because the claims do not recite additional elements that amount to a technological improvement (see MPEP 2106.05(a)). The additional elements such as “sensor,” and “communication network,” under their broadest reasonable interpretation, are generic components that perform routine, and well-understood functions such as data collection, analysis and transmitting results (see MPEP 2106.05(d)). The additional limitations merely add instructions to implement the abstract idea on a computer or use the computer as a tool to perform the abstract idea and/or link the use of the judicial exception to a particular technological environment or field of use (see MPEP 2106.05(h)). Therefore, the claims are directed to an abstract idea.
[STEP2B] The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements such as “sensor,” “DRL algorithm” and “artificial intelligence systems,” are under their broadest reasonable interpretation, conventional tools for implementing the abstract idea. The specification discloses that the first, second, and the third AI systems receive and process data, assign labels, provide corrective suggestions (Par. [0026]), and that the system can be deployed on an electronic device or cloud application (Par. [0028]). The recitation of these additional elements does not amount to significantly more than the judicial exception, but rather link the abstract idea to a particular technological environment or field of use. Therefore, claim(s) 1-17 are not directed to eligible subject matter but are found to recite a judicial exception without significantly more.
Claim(s) 2-10, 12-16 is/are dependent on supra claim(s) and include all the limitations of the claim(s). Therefore, the dependent claim(s) recite(s) the same abstract idea. The claim recites no additional limitations. For example, claim(s) 2 and 8 further define the type of AI algorithms used, claim(s) 3-4 and 9 further define the types of data sources and sensors used to collect information, claim(s) 6-7 further recites a user interface for displaying results, claim 12 further recites mathematical processing using Bayesian logic, and claim(s) 13-16 further define data acquisition and evaluation.
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea and the claim is therefore directed to the judicial exception. Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Thus, even when viewed as a whole, nothing in the claim adds significantly more (i.e., an inventive concept) to the abstract idea.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1-6, 9-10, and 13-16 is/are rejected under 35 U.S.C. 102(a)(1) and (a)(2) as being anticipated by May et al. (US 20190009133 A1) hereinafter referred to as May.
In regards to claim(s) 1, May teaches a system for intelligent physical event analysis and providing individualized assessment (Abstract and Par. [0017]; “The systems may use movement skill assessment and diagnostics at distinct levels of the human movement system hierarchy to specify training goals,” and Par. [0126]), comprising:
a source for capturing data of an event, wherein the data of the event is categorized as environmental feedback or input data (Par. [0851]; “a sensor system comprising one or more sensors configured to obtain movement data for a subject performing an activity”, and Par. [0017]-[0019] teach a system that captures movement related data from interactions between a player, environment, and equipment);
a first artificial intelligence (AI) system configured to receive the environmental feedback from one or more sensors, wherein the first AI system incorporates multiple AI algorithms to consolidate sensor data and determine the event (Par. [0120]; “machine learning techniques can be applied to analyze the distribution of features and characteristics of the movement, as well as to aggregate and classify the data”);
wherein an event label is assigned to each event by the first AI system ([0139]; “FIG. 6 gives an overview of the movement processing starting from the extraction of movement units, their classification”);
a second AI system configured to process the input data in conjunction with the event labels provided by the first AI system;
wherein the second AI system is configured to analyze the input data leading to the event and provide a suggestion for a corrective action (Par. [0115] teaches a cueing system that processes movement data in combination with a user’s current movement state and analyzes the data to determine and generate feedback cues (e.g., “The cue generator translates cue signals into physical stimuli; the system operates in real-time to provide feedback as the user participates in an activity.”));
a third AI system configured to use captured data to predict expected outcomes by selecting the event labels based on the input data, and comparing a prediction with the suggestion for the corrective action (Par. [0112] and FIG. 21; “Session data 226 can be provided to the extractor 201. The extractor 201 output generates a motion model 205 which can then be used for skill assessment and diagnostics 206 based on reference skill data 207,” and “collects information from user training or play,” and “a feedback loop between a movement process 222 and a cueing system”),
thereby facilitating iterative updates across the first and second AI systems to enhance predictive accuracy, wherein once trained, the third AI system is configured to autonomously provide coaching advice based on input data (Par. [0114] teaches the three primary multiple loop systems (e.g., “Assessment Loop (AL), Training Loop (LP), Feedback Augmentation Loop (FL)”) that continuously update training and feedback);
In regards to claim(s) 2, May further teaches the multiple AI algorithms selected from the group consisting of Bayesian learning, and machine learning (Par. [0527] teaches the use of a Bayesian Network and an inference algorithm to analyze observations and generate a diagnostic (e.g., “An inference algorithm uses the Bayesian Network and the observations to determine the most likely explanation for the observations, i.e., diagnostic.”)).
In regards to claim(s) 3, May teaches the environmental feedback includes data sources selected from the group consisting of radar devices, accelerometers, sports capture devices, GPS watches, basketball shot trackers, swimming lap counters, smart balls, drone cameras, and sport sensors. (Par. [0117] teaches the use of devices such as a “smart phone” and other “motion tracking”)
In regards to claim(s) 4, May teaches the input data comprises data captured from sources including, but not limited to, video feeds from cameras, audio data, wearable fitness trackers, smartwatch data, and wearable biomechanical sensors (Par. [0627] further teaches IMU, visual or optical tracking systems, video cameras, and vision processing for extracting motion data).
In regards to claim(s) 5, May further teaches the system further comprising a cloud application configured to store, process, and manage data collected from the first, second and third AI systems, and to provide accessibility to the system across different devices and platforms (Par. [0118]; “The processing is distributed across typical internet of things (IoT) components, such as the wearable/embeddable devices, smart devices and cloud infrastructure.”).
In regards to claim(s) 6, May further teaches the system assigns multiple event labels simultaneously depending on the complexity of the event data received from the environmental sensors (Par. [0139] and FIG. 6 teach classification of movement units; “extraction of movement units, their classification”).
In regards to claim(s) 9, May teaches the system, further comprising the one or more sensors configured to detect a physical training environment (Par. [0851]; “a sensor system comprising one or more sensors configured to obtain movement data for a subject performing an activity”).
In regards to claim(s) 10, May teaches the system further includes user interfaces designed to display the analyzed data and suggested corrective actions in a user-friendly format (Par. [0062] teaches a skill status display; e.g., “FIG. 44A is an illustration of a skill status screen showing the skill elements arranged according to their acquisition stage (Patterns to Form, Patterns to Consolidate, and Patterns to Optimize)”).
In regards to claim(s) 13, May teaches the image feed is a camera (Par. [0627]; “Examples include the use of video cameras 70 that capture the broader agent behavior and the task environment 50.”).
In regards to claim(s) 14, May teaches the system and method further comprising receiving a pre-existing video (Par. [0628]; “video cameras on the subjects and/or environment make it possible to determine which elements or events the performers are attending to at any given time”).
In regards to claim(s) 15, May teaches the system further comprising assessing and assigning a skill level to the target player (Par. [0440], [0062] and [0851] teach generating performer profiles based on performance data, organizing skill elements by acquisition stages, and defining skill levels based on individual and population data).
In regards to claim(s) 16. May teaches the system further comprising storing the skill level of the target player (Par. [0043] and FIG. 25 teaches collecting and processing activity data over time (“During each session, activity data is collected and processed.”) and Par. [0053] teaches tracking performance trends over a period of time (“FIG. 35 is an illustration of play activity summation over a calendar period showing sets and sessions.”)).
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.
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.
Claim(s) 7-8, 11-12, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over May et al. (US 20190009133 A1) hereinafter referred to as May, in view of Lillicrap et al. (US 20170024643 A1) hereinafter Lillicrap, further in view of Kashyap (WO 2022226365 A1), and further in view of Trehan et al. (US 20220023718 A1) hereinafter Trehan.
In regards to claim(s) 7, May further teaches the system is adapted to provide real-time coaching to a user based on an analysis of the actions of the user and the event data (Par. [0115]; “the system operates in real-time to provide feedback as the user participates in an activity.”), wherein the real-time coaching is personalized to the user’s specific characteristics (Par. [0115]; “The form of transducer is determined by the…user characteristics, equipment parameters, environment status, and/or other concerns.” )
May does not explicitly teach wherein the real-time coaching is personalized to the user’s specific previous performance, but instead discloses the feedback cue signal is provided in real time to the user (Par. [0869]-[0870]). However, Trehan teaches providing feedback after completion of an activity (Par. [0050]; “the feedback may be provided after the user has completed the given activity.”).
It would have been obvious to one of ordinary skill in the art at the time of the claimed invention to modify May’s real-time feedback system to further incorporate Trehan’s teaching of providing feedback after completing an activity. A person of ordinary skill in the art would have been motivated to combine Trehan’s system of providing post activity feedback with May’s teachings of real-time feedback to allow the system to provide personalized coaching advice based not only on current user characteristics but also on previous performance.
In regards to claim(s) 8, May does not explicitly teach the third AI system includes a predictive model that employs deep reinforcement learning algorithms to independently generate a prediction and coaching advice without requiring real-time data input from the first AI system. However, Lillicrap teaches a deep reinforcement learning including an actor neural network, that once trained, maps observations to actions using learned parameters (Par. [0021]-[0022]; “the reinforcement learning system 100 trains the actor neural network 110 to determine trained values of the parameters ”), (Par. [0022]; “the reinforcement learning system 100 can process the observation using the actor neural network 110 to map the observation to a new action in accordance with the trained values of the parameters” and Par. [0057]; “By repeatedly performing the process 300 on multiple different minibatches of experience tuples, the system can train the actor neural network to determine trained values of the parameters”). Under the broadest reasonable interpretation, generating actions from input observations using learned parameters corresponds to independently generating predications and coaching advice based on input data.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify May’s data-driven closed-loop training system to include the reinforcement learning techniques taught by Lillicrap, such that the predictive model of May employs deep reinforcement learning algorithms capable of mapping observations to actions and generating predictions and coaching advice without requiring real-time input from the first AI system. A person of ordinary skill in the art would have been motivated to incorporate Lillicrap’s reinforcement framework into May’s system because Lillicrap, explicitly teaches that, once trained, the actor neural network can operate based on learned parameters to determine actions from observations, thereby reducing reliance on continuous real-time data streams.
In regards to claim(s) 11 and 17, May teaches the system further comprising:
an electronic device operably connected to the first, second and third AI system (Par. [0039] and FIG 22 teaches a data driven closed-loop training system organized around three primary feedback loops),
the electronic device configured to transmit and receive the captured data over a communication network (Par. [0172] and FIG. 22 teaches an augmentation system architecture connecting the AI components through communications and UI systems over a network; “the human system augmentation loop (with communications and UI systems)”);
wherein the first, second and third AI system is associated with a target player (Par. [0440] and FIG. 29 teaches generating performer profiles based on performance profiles; “The information extracted from the population analysis makes it possible to determine performer profiles.”);
a video feed operably connected to the first AI system, wherein the third AI system is adapted to predict the consequences of the actions of the target player (Par. [0627] teaches a video feed connected to the AI processing system whereby video cameras capture “agent behavior and task environment”);
a sensor operably connected to the video feed or the second AI system, the sensor configured to measure a physical feedback; an action set comprising a plurality of training improvement categories (Par. [0851]-[0853] teaches collecting movement data from a variety of sensors (e.g., “one or more sensors”), including a “video sensor”, identifying and analyzing movement patterns to identify one or more skill attributes associated with a performed activity, and using those attributes to assess performance and determine targeted training goals (e.g., “derive one or more training elements”));
a Bayesian filter comprising a Bayesian logic, wherein the Bayesian filter is operably connected to the third AI system and configured to receive the physical feedback from the sensor; wherein the first, second and third AI system are configured to perform a sports training improvement method (Par. [0527]-[0529] and FIG. 27-28 teaches a Bayesian filter comprising Bayesian logic through a diagnostic system combining knowledge representation, observations, and an interface mechanism that Par. [0529]; e.g., “The diagram can be structured as a Bayesian Belief Network and used as part of the diagnostic system.”). May further teaches that the Bayesian Belief Network receives physical observations corresponding to sensor-derived metrics across the movement system hierarchy (Par. [0528]), the method comprising:
receiving the video feed, wherein the target player is identifiable within the video feed (Par. [0627; “Vision processing can also be used to extract information about the motion of individual body segments 15.”]);
receiving the physical feedback from the sensor (Par. [0851]; “receive the movement data from the one or more sensors, wherein the subject performs a primary movement unit associated with the activity”);
providing the training improvement category to the target player in real-time and presenting the training improvement category as the coaching advice (Par [0115]; “the system operates in real-time to provide feedback as the user participates in an activity.”);
applying the Bayesian filter to convert the physical feedback to a statistical measurement (Par. [0527] and FIG. 27; “An inference algorithm uses the Bayesian Network and the observations to determine the most likely explanation for the observations, i.e., diagnostic.”);
comparing the training improvement category predicted by the neural network to the statistical measurement generated by the Bayesian filter (FIG. 21 and FIG. 22 teach a closed-loop training system whereby outputs of predictive processes are evaluated against diagnostic results and repeatedly refined over time);
May does not explicitly teach processing the video feed and the physical feedback using multiple convolutional layers. However, Kashyap teaches a neural network architecture that processes image and video data using multiple convolutional layers (Par. [70]; “The neural network 300 receives an image 310 and utilizes a U-Net style keypoint detector 320 (or other convolutional neural network”) and (Par. [81] teaches processing a set of “images or video stream” frames through “the neural network framework… which utilizes keypoint detection techniques.”).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the multi-layer convolutional architecture taught by Kashyap in order to process the video feed and physical feedback data. A person of ordinary skill in the art would have been motivated to make this modification to improve the accuracy and efficiency of feature extraction from video and sensor data.
May does not explicitly teach predicting a training improvement category from the action set and sorting the video feed into the training improvement category via full connected layers and a long short term memory layer. Instead, May discloses a data driven training system that categorizes movement performance deficiencies into actionable training improvement categories and utilizes diagnostic results to provide real-time corrective recommendations to the user (Par. [0531]; “Information from the diagnostic system, when applied to the larger control hierarchy, can also be used to analyze games or task performance and even be used in real-time to recommend actions”).
However, Kashyap teaches a system that processes video input and classifies it using a neural network including fully connected layers (Par. [81]-[88] teach an exercise detection system that sorts a video feed into a predicted exercise classification via fully connected layers and an LSTM layer operating on video frame sequences; “The LSTM layer 745, via the applied techniques, outputs an exercise classification 748 for the exercise depicted in the images 710.”). It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the classification output of Kashyap to implement in the training improvement categories taught by May. A person of ordinary skill in the art would have been motivated to incorporate Kashyap’s machine learning classification pipeline into May’s system to enable the system to translate exercise classifications into coaching advice.
May does not explicitly teach analyzing the statistical measurement produced by the Bayesian filter via a DRL algorithm, wherein the DRL algorithm is configured to learn through trial and error to provide better feedback and make better decisions. However, Lillicrap teaches a reinforcement learning system where an actor neural network maps the observation to a new action and directs the agent accordingly (Par. [0022]; “the reinforcement learning system… map the observation to a new action in accordance with the trained values of the parameters of the actor neural network”).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lillicrap’s DRL system into May’s Bayesian diagnostic system to improve adaptive decision making. A person of ordinary skill in the art would recognize that the performance data generated by May could be used as input to Lillicrap’s reinforcement learning model, allowing the system to learn improved feedback strategies.
May does not explicitly teach updating a weight of the first, second and third AI system. However, Lillicrap teaches a system that updates current values of the parameters of the critic and actor neural networks (Par. [0050]-[0055] and Par. [0033]; “During the training, the reinforcement learning system 100 also periodically updates the values of the parameters of the target critic neural network 160 and the values of the parameters of the target actor neural network 150…the values of the parameters of the critic neural network 140”).
It would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify May’s system to include the parameter updating techniques taught by Lillicrap so that the first, second, and third AI systems are trained by periodic updates to their respective model parameters. A person of ordinary skill in the art would have been motivated to combine these teachings because updating model weights during training is a well-known aspect of machine learning systems to improve model accuracy over time.
In regards to claim(s) 12, May teaches wherein the Bayesian logic is configured to perform a method of identifying the playing error cause of the target player (Par. [0526]; “Bayesian inference, from observations to determine probable causes of specific phenomena.”) and (Par. [0527]; “An inference algorithm uses the Bayesian Network and the observations to determine the most likely explanation for the observations”), the method of identifying the playing error cause comprising the following steps:
creating or receiving a statistical profile of the target player (Par. [0708]; “Skill Assessment and Diagnosis 130 represents the processes used to determine the parameters that characterize a subject's skill elements, which subsequently determine a subject's skill profile and skill status”);
receiving the physical feedback from the sensor (Par. [0121]; “one or more motion sensors, either embedded or deployed in the user's environment, can be used with the system to provide measurements of movement dynamics”);
extracting relevant information from the physical feedback (Par. [0533]; “The main components of such as DNN may include…learning the movement and broader performance features (conditions and contingencies associated with contextual details)”);
converting the relevant physical feedback and the statistical profile into a suitable format to be processed by the Bayesian filter ((Par. [0529] teaches structuring physical observations (e.g., motion and technique) as a Bayesian Belief Network).);
training the Bayesian filter using the statistical profile and relevant physical feedback collected via updating the parameters of the Bayesian filter to fit the statistical profile and relevant physical feedback (Par. [0530]; “The diagnostic system can combine expert knowledge…with detailed movement functional analysis and direct diagnostics based on assessments.”);
applying the Bayesian filter to convert the physical feedback to a measurement and determine the most likely cause of the playing error of the target player via calculating the posterior probabilities of each potential sport error cause given the observed data and selecting the playing error cause with the highest probability (Par. [0526]-[0527] teach applying a Bayesian inference model to convert observed physical feedback into probabilistic measurements and determine the most likely playing error cause, “determine the most likely explanation for the observations, i.e., diagnostic” );
sending the playing error cause predicted by the Bayesian filter to be processed by the DRL algorithm (FIG. 27 and Par. [0531]; “Information from the diagnostic system, when applied to the larger control hierarchy, can also be used to analyze games or task performance”).
May does not explicitly teach generating a new data point via the relevant physical feedback. However, Lillicrap teaches a reinforcement learning system that generates an experience tuple and stores the generated data for use in training neural networks (Par. [0044]; “The system generates an experience tuple that includes the current observation, the selected action… and the next observation and stores the generated experience tuple in a replay memory for use in training the actor neural network (step 210).”). May discloses a movement assessment system that receives and processes physical feedback data from motion sensors (Par. [0121]) and utilizes data driven feedback loops to iteratively improve training outcomes (Par. [0178]-[0180]; “The framework describes the management of data sets used to support skill assessment and diagnostics”).
It would have been obvious to a person of ordinary skill in the art before the effective filling date of the claimed invention to modify May’s system to incorporate Lillicrap’s experience tuple generation mechanism so that new data points are generated from physical feedback received by May’s motion sensors. A person of ordinary skill in the art would have been motivated to make this modification of incorporating Lillicrap’s technique to enable May’s system to systematically generate structured training data from observes interactions, thereby enhancing the ability to learn from user performance over time.
Conclusion
Accordingly claim(s) 1-17 are rejected.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BUSHIRA MUSA whose telephone number is (571)272-9156.
The examiner can normally be reached Mon-Fri 7:30am-4:30pm.
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, Kang Hu can be reached at 5712701344. 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.
/BUSHIRA MUSA/Examiner, Art Unit 3715
/KANG HU/Supervisory Patent Examiner, Art Unit 3715
Prosecution Insights