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 .
Herein after “it would have been obvious” should be read as “it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention”.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-20 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.
Long et al teaches a generic user selecting an AI model as originally claimed. Long et al does not teach the newly added limitations of determining a user profile from user inputs, then selecting an AI model, from the user profile. Long et al also does not teach the common limitation of providing the processed user input to an application. Therefore, references are being cited that teach these newly claimed limitations.
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.
Claim(s) 1-5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, and Aurongzeb et al PN 2016/0044422.
In regards to claim 1: Long et al teaches a Human Interface Device (HID) ([0177] "FIG. 17 shows an example user interface 1700 for selecting and deploying AI models on the hybrid cloud-edge platform"), comprising: a processor ([0022] "According to an aspect of the present disclosure, a non-transitory computer-readable storage medium includes instructions that, when executed by a processor, cause a coordinator cluster of a cloud-edge computing platform to receive a data file associated with a workload"); and a memory coupled to the processor ("a non-transitory computer- readable storage medium"), the memory having program instructions stored thereon ("includes instructions that, when executed by a processor,") that, upon execution, cause the HID to: receive an Artificial Intelligence (AI) model selection from the user ([0177] "FIG. 17 shows an example user interface 1700 for selecting and deploying AI models on the hybrid cloud-edge platform"); and in response to the AI model selection, load a corresponding AI model ([0177] "FIG. 17 shows an example user interface 1700 for selecting and deploying AI models on the hybrid cloud-edge platform") configured to receive raw user input ([0225] "In some embodiments, user computing entity 2300 may include a user interface, comprising an input interface 2350 and an output interface 2352, each coupled to processing unit 2310. User input interface 2350 may comprise any of a number of devices or interfaces allowing the user computing entity 2300 to receive data, such as a keypad (hard or soft), a touch display, a mic for voice/speech, and a camera for motion or posture interfaces" [0212] "During model training 2230, pre-processing involves cleaning, normalizing, and transforming raw data into a format suitable for learning patterns") and to produce processed user input ([0078] "In certain respects, the disclosed systems can feature a pipeline computational architecture and method wherein the data to be processed can be fed into a first computation model (e.g., a first machine learning model) on a first network type (e.g., cloud) and then the output can be fed into a second computation model (e.g., a second machine learning model) on a second network type (e.g., THETA Edge Network). Thus, the disclosed systems can perform computations for a workload using multiple models (e.g., deep-learning models or video processing models) consecutively"). Long does not teach the collected raw user input to “determine, based on the collected raw user input, a user profile”, using the user profile to select an AI model or expressly supplying the user inputs to an application. Kwatra et al teaches “[0057] “This determination can be made by monitoring feedback across a large set of users, with varying characteristics (age, geography, etc). In such scenarios, the voice response system creates separate AI models for specific user profiles (by geography, age, etc.) and maintains those for specific user profiles while still maintaining the original models for general responses”. It would have been obvious to select the AI models based on a user profile because this would have been obvious to use a user profile to select/create an AI model because this would have allowed the AI model to be consistent to user preferences. Price et al teaches abstract “A system for receiving and presenting content at a user location includes a profiling agent that creates a user profile based upon a plurality of inputs by a user. The user profile represents preference characteristics of the user”. Price et al also gives channel selection as an example of the user inputs thus the user input is raw user inputs. Column 5 line 19 et. seq. “In the illustrated embodiment, the list 44 includes the contents, the days and the time of days. This user prefers golf and tennis and watches mainly on a Saturday and on a Sunday between 7 8 a.m. and between 8 11 p.m. The list 44 is based upon inputs by the user, either by tracking the user's selected channels or by direct input of a preference characteristic into the list 44 by the user”. It would have been obvious to use the collected user inputs to create user profiled because this would have alleviated a user from having to create his/her own profile. While Price et al states the user input is a channel selection Price never expressly states the channel selection input is provided to an app to actually select a channel. Aurongzeb et al teaches [0014] “In one scenario, user input is provided to a word processing program on an information handling system such as a tablet”). It would have been obvious to provide the user input to an application to which it is directed because this would have prevented never using the user inputs.
In regards to claim 2: Long et al teaches the user interface device may be a mouse or keyboard. ([0237] "For interface with a user, the hardware may include one or more user input devices (e.g., a keyboard, a mouse, a scanner, a microphone, a camera, etc.) and a display (e.g., a Liquid Crystal Display (LCD) panel)").
In regards to claim 3: Long et al. teaches a keypad. ([0225] "User input interface 2350 may comprise any of a number of devices or interfaces allowing the user computing entity 2300 to receive data, such as a keypad (hard or soft), a touch display, a mic for voice/speech, and a camera for motion or posture interfaces").
In regards to claim 4: Long et al teaches selecting different AI models such as Stable Diffusion, Llama 2, and other off-the-shelf models with clicks which is done by a button ([0177] "FIG. 17 shows an example user interface 1700 for selecting and deploying AI models on the hybrid cloud-edge platform. AI developers using the hybrid cloud-edge computing platform can easily select and deploy popular Al models such as Stable Diffusion, Llama 2, and other off-the-shelf models with just a few clicks and build their AI powered apps on top. These Al powered apps can be configured to be trained and deployed efficiently on the EdgeCloud platform").
In regards to claim 5: Long et al teaches a language model. ([0084] "In some embodiments, the disclosed cloud-edge computing platform runs AI inference applications that include but are not limited to generative text-to-image, text-to-video, large language models (LLM), and other custom models").
Claim(s) 6-7 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, and Aurongzeb et al PN 2016/0044422 as applied to claim 1 above, and further in view of Schroyer et al PN 6,571,299.
In regards to claim 6: Long et al teaches the input device may be a scanner or a keyboard. Long et al does not expressly teach the data is a "scancode". Schroyer et al teaches the data from a keyboard is a scancode. (Column 1 line 60 et. seq. 'The hardware device, connected between the standard keyboard and the keyboard port, that recognizes and accepts ID-Codes from transmitters with which it has previously been trained to recognize, associates each ID-Code with the scancode, or modified scancode, learned in the training sequence and held in its memory, and sends that scancode on to the CPU, the same as if it were coming from the standard keyboard'). It would have been obvious to have the user input be a scancode because this is the form of data from a scanner and keyboard.
In regards to claim 7: Schroyer et al teaches a modified scancode ("modified scancode").
Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, and Aurongzeb et al PN 2016/0044422 as applied to claim 1 above, and further in view of Jorasch et al PN 2021/0342020.
In regards to claim 8: Long et al teaches the input being a mouse or a keyboard but does not teach a sensor or measuring click/press pressure/frequency/velocity Jorasch et al teaches (1471] "Utilizing biometric inputs from the devices, an AI module could be trained that analyzes physical and mental performance aspects of game play. For example, the module might detect that a player performs poorly in a given match and the player had a slight hand tremor as measured by an EMG sensor or inferred from mouse or keyboard pressure"). It would have been obvious to allow biometric inputs such as pressure or frequency as the input data because this would have prevented limiting the types of analysis performed.
Claim(s) 9-10, 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, Aurongzeb et al PN 2016/0044422, and Jorasch et al PN 2021/0342020 as applied to claim 8 above, and further in view of Carreio et al PN 10,382,898.
In regards to claims 9-10, 14: Long et al teaches loading different Al models. Carreio et al teaches user profiles including biometric information (Column 34 line 1 et, seq. "personal or biometric information of a user for user-authentication or experience-personalization purposes" "user to provide a reference image (e.g., a facial profile, a retinal scan)") and loading an AI model based upon whether or not a profile matches. (Column 24 line 59 et. seq. "Additionally, or alternatively, the localization application may implement a first machine learning model for a first set of users with shared characteristics and may implement a second machine learning model for a second set of users with a different set of shared characteristics"). It would have been obvious to use biometrics to select the AI model because this would have allowed multiple different people to have different configurations.
Claim(s) 11-13, 18-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, and Aurongzeb et al PN 2016/0044422 as applied to claim 1 above, and further in view of Spangler et al PN 2015/0199028 and Hamlin et al PN 2025/0306956.
In regards to claims 11-12, 20: Long et al is only directed to a user input selecting an Al model from a user input device. Long et al is silent upon an OS agent providing the AI models or there being an embedded controller where the data path is with or without involvement of the embedded controller. Hamlin et al teaches an operating system agent providing the AI model. ([0212] "In some cases, BKC service 602N and/or an OS agent may execute or trigger the execution of AI/ML models trained with node setting information and OS setting information sufficient to infer a suitable BKC"). It would have been obvious to have the operating system provide the AI models because this would have given a location source for the AI model(s). Spangler et al teaches two paths for the keyboard to interact with the processor one without involvement with an embedded controller and one with involvement of an embedded controller ([0016] "FIG. 5B shows the processor interfacing with the keyboard through the embedded controller and a communication link that bypasses the embedded controller according to an aspect of the subject technology". See also figure 5B). It would have been obvious to either involve the embedded controller or to bypass the embedded controller because this would have prevented limiting the possible connection of the user interface device.
In regards to claims 13, 18: Long et al is only directed to a user input selecting an AI model from a user input device. Long et al is silent upon an OS agent providing the AI models Hamlin et al teaches an operating system agent providing the AI model. ([0212] "In some cases, BKC service 602N and/or an OS agent may execute or trigger the execution of AI/ML models trained with node setting information and OS setting information sufficient to infer a suitable BKC"). It would have been obvious to have the operating system provide the AI models because this would have given a location source for the Al model(s).
In regards to claim 19: Long et al teaches a language model. ([0084] "In some embodiments, the disclosed cloud-edge computing platform runs AI inference applications that include but are not limited to generative text-to-image, text-to-video, large language models (LLM), and other custom models").
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, Aurongzeb et al PN 2016/0044422 and Carreio et al PN 10,382,898.as applied to claim 14 above, and further in view of Jorasch et al PN 2021/0342020.
In regards to claim 15: Long et al teaches the input being a mouse or a keyboard but does not teach a sensor or measuring click/press pressure/frequency/velocity Jorasch et al teaches (1471] "Utilizing biometric inputs from the devices, an AI module could be trained that analyzes physical and mental performance aspects of game play. For example, the module might detect that a player performs poorly in a given match and the player had a slight hand tremor as measured by an EMG sensor or inferred from mouse or keyboard pressure"). It would have been obvious to allow biometric inputs such as pressure or frequency as the input data because this would have prevented limiting the types of analysis performed.
Claim(s) 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, Aurongzeb et al PN 2016/0044422 and Carreio et al PN 10,382,898. as applied to claim 14 above, and further in view of Hamlin et al PN 2025/0306956.
In regards to claim 16: Long et al is only directed to a user input selecting an AI model from a user input device. Long et al is silent upon an OS agent providing the AI models Hamlin et al teaches an operating system agent providing the AI model. ([0212] "In some cases, BKC service 602N and/or an OS agent may execute or trigger the execution of AI/ML models trained with node setting information and OS setting information sufficient to infer a suitable BKC"). It would have been obvious to have the operating system provide the Al models because this would have given a location source for the AI model(s).
Claim(s) 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Long et al PN 2025/0123902 in view of Price et al PN 6,986,154, Kwatra et al PN 2022/0172714, Aurongzeb et al PN 2016/0044422 and Carreio et al PN 10,382,898 as applied to claim 14 above, and further in view of Lee PN 6,181,325.
In regards to claim 17: Long et al teaches using the HID/user interface to select the AI model however never mentions controlling the user interface device. Lee teaches (column 1 lines 33 et. seq. "To control movement of the mouse pointer precisely in the notebook computer, either changing the mouse resolution to higher one or adjusting the mouse pointer speed slowly is needed. The adjustment of the pointer speed should be performed through the corresponding control program each time it is required to fit for the use"). It would have been obvious to use a control application/program to such as mouse.exe to control the HID/user interface device because this would have allowed a user to adjust features of the HID such as pointer speed or keyboard repeat rate.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
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 PAUL R MYERS whose telephone number is (571)272-3639. The examiner can normally be reached telework M-F start 7-8 leave 4-5.
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, Jaweed Abbaszadeh can be reached at 571-270-1640. 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.
/Paul R. MYERS/Primary Examiner, Art Unit 2176
Prosecution Insights