DETAILED ACTION
This is in response to US App. 18/859,849. Claims 44-66 have been examined. Claims 1-43 have been cancelled.
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 . 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.
Claim Rejections - 35 USC § 102
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.
Claim(s) 44-66 are rejected under 35 U.S.C. 102(a)(1) as being unpatentable by Pantelidou et al. (WO 2022/008037; included in IDS; hereafter WO8037).
Regarding Claim 44,
A method performed by a wireless device for fallback operation of a machine learning (ML) model, the method comprising:
transmitting a message indicating a capability of the wireless device for supporting a combination of at least one ML-based feature for a functionality and at least one fallback feature for the functionality to a network node [WO8037: p. 9 / lines 30-33; it is assumed that the 5G UE and gNB are capable of operating with a support of ML models; one or more trained ML models may be available at the UE to perform a certain task/to solve a certain problem; the UE is also allowed to use a non-ML algorithm to perform the given task/solve the given problem; p. 10 / lines 7-10; the UE indicates to the network its ability to execute an ML model a given time; the ML model execution related ability is different from existing static UE capabilities since it can vary with time and depends not only on UE type, but also on the current UE ability (power, memory, etc.); additionally, the UE may indicate to the network a Default Behavior to which it falls back when UE is not able to execute and/or train the ML model. The Default Behavior may also depend on UE type and ability (power, memory, etc.)];
operating the at least one ML-based feature for the functionality; and operating the at least one fallback feature for the functionality [WO8037: p. 12 / lines 29-31; in some example embodiments, the UE may indicate to the network its Default Behavior associated to an ML model (that is related to a certain optimization problem), to which it falls back when ML execution and/or training is not possible (UE is unable for ML); p. 13 / lines 21-22; as shown in Fig. 4, network (gNB) sends to UE a message “Activate ML model” with which network activates a ML model (e.g., ML model m) to solve a certain problem pm; p. 14 / lines 30-36; Fig. 6 shows an example where the network detects that the current ML model used by the UE is suboptimal; this can be the case if the network observes that the current ML model does not perform well, for example if the network conditions have changed; in this case, as shown in Fig. 6, the network upon detection of suboptimal operation of ML model m for a given problem pm, sends a De-Activate ML model message to the UE; the network may signal to the UE to De-Activate multiple ML models related to different problems; the UE receiving the De-Activate ML model message reverts to Default Behavior for all the indicated problems …].
Regarding Claim 45,
A wireless device capable of fallback operation of a machine learning (ML) model, the wireless device comprising processing circuitry operable to:
transmit a message indicating a capability of the wireless device for supporting a combination of at least one ML-based feature for a functionality and at least one fallback feature for the functionality to a network node [WO8037: p. 9 / lines 30-33; it is assumed that the 5G UE and gNB are capable of operating with a support of ML models; one or more trained ML models may be available at the UE to perform a certain task/to solve a certain problem; the UE is also allowed to use a non-ML algorithm to perform the given task/solve the given problem; p. 10 / lines 7-10; the UE indicates to the network its ability to execute an ML model a given time; the ML model execution related ability is different from existing static UE capabilities since it can vary with time and depends not only on UE type, but also on the current UE ability (power, memory, etc.); additionally, the UE may indicate to the network a Default Behavior to which it falls back when UE is not able to execute and/or train the ML model. The Default Behavior may also depend on UE type and ability (power, memory, etc.)];
operate the at least one ML-based feature for the functionality; and operate the at least one fallback feature for the functionality [WO8037: p. 12 / lines 29-31; in some example embodiments, the UE may indicate to the network its Default Behavior associated to an ML model (that is related to a certain optimization problem), to which it falls back when ML execution and/or training is not possible (UE is unable for ML); p. 13 / lines 21-22; as shown in Fig. 4, network (gNB) sends to UE a message “Activate ML model” with which network activates a ML model (e.g., ML model m) to solve a certain problem pm; p. 14 / lines 30-36; Fig. 6 shows an example where the network detects that the current ML model used by the UE is suboptimal; this can be the case if the network observes that the current ML model does not perform well, for example if the network conditions have changed; in this case, as shown in Fig. 6, the network upon detection of suboptimal operation of ML model m for a given problem pm, sends a De-Activate ML model message to the UE; the network may signal to the UE to De-Activate multiple ML models related to different problems; the UE receiving the De-Activate ML model message reverts to Default Behavior for all the indicated problems …].
Regarding Claim 46,
wherein the at least one ML-based feature is based on one ML model that is split in two parts, with one part located at the wireless device and the other part located at the network node [WO8037: p. 3 / line 20; the ML model execution may be at the UE side, at the network side, or in both].
Regarding Claim 47,
wherein the at least one fallback feature is a feature that fulfills comparable functionalities as the ML-based feature, but is not preferred compared to the ML-based feature [WO8037: p. 7; lines 29-30; UE may fall back to a Default Behavior if it is not able to execute and/or train a ML model because of its current state; p. 12 / lines 1-7; in line with the change of the “ML State Indication” value, the UE may also update its Default Behavior for a given task/problem and inform the network thereabout; the Default Behavior may not be a unique behavior through the course of time of UE operation and may depend on the UE state; for instance, a UE, at times when its memory is full, can indicate to the network that its default behaviour is to “run non-ML algorithm” for a certain task but if later on in time its state changes it can indicate that it is ready to “run light ML algorithm” for the task].
Note:
A fallback is always a back stop (i.e. a non-preferred option).
Regarding Claim 48,
wherein the at least one fallback feature is a feature that has higher capabilities than the ML-based feature, but the fallback feature is not preferred [WO8037: p. 12 / lines 1-7; in line with the change of the “ML State Indication” value, the UE may also update its Default Behavior for a given task/problem and inform the network thereabout; the Default Behavior may not be a unique behavior through the course of time of UE operation and may depend on the UE state; for instance, a UE, at times when its memory is full, can indicate to the network that its default behaviour is to “run non-ML algorithm” for a certain task but if later on in time its state changes it can indicate that it is ready to “run light ML algorithm” for the task].
Note:
A non-ML algorithm has the capability to execute when a memory is full but a ML algorithm does not.
Regarding Claim 49,
wherein the at least one fallback feature is based on a non-ML-based algorithm [WO8037: p. 9 / lines 30-33; it is assumed that the 5G UE and gNB are capable of operating with a support of ML models; one or more trained ML models may be available at the UE to perform a certain task/to solve a certain problem; the UE is also allowed to use a non-ML algorithm to perform the given task/solve the given problem.
Regarding Claim 50,
wherein the at least one fallback feature is a ML-based algorithm [WO8037: p. 12 / lines 1-7; in line with the change of the “ML State Indication” value, the UE may also update its Default Behavior for a given task/problem and inform the network thereabout; the Default Behavior may not be a unique behavior through the course of time of UE operation and may depend on the UE state; for instance, a UE, at times when its memory is full, can indicate to the network that its default behaviour is to “run non-ML algorithm” for a certain task but if later on in time its state changes it can indicate that it is ready to “run light ML algorithm” for the task].
Regarding Claim 51,
wherein the message indicates whether the at least one fallback feature and the at least one ML-based feature may be executed simultaneously [WO8037; p. 11 / lines 1-2; Option a: re-use existing UE Capability IE together with a new IE (e.g. “ML state indication”)].
Regarding Claim 52,
the processing circuitry further operable to receive a first configuration message that configures the wireless device to operate the at least one ML-based feature [WO8037: p. 13 / lines 21-22; as shown in Fig. 4, network (gNB) sends to UE a message “Activate ML model” with which network activates a ML model (e.g., ML model m) to solve a certain problem pm].
Regarding Claim 53,
the processing circuitry further operable to receive a first configuration message that configures the wireless device to operate the at least one ML-based feature and at least a fallback feature simultaneously [WO8037; p. 11 / lines 1-2; Option a: re-use existing UE Capability IE together with a new IE (e.g. “ML state indication”)].
Regarding Claim 54,
the processing circuitry further operable to receive a second configuration message that configures the wireless device to deactivate the at least one ML-based feature and activate the at least one fallback feature [WO8037; p. 14 / lines 14-17, 19-26; at some point in time, UE detects a State Change that affects its ability to execute and/or train the ML model; in this situation, the UE can declare to the network it is not able for full ML processing (for instance using Option a or Option b), and UE autonomously falls back to Default Behavior; in the example of Fig. 5, activation of an ML model and detecting the UE state change are the same as in the example of Fig. 4. However, differently than in Fig. 4, if the UE detects its state change (inability to execute and/or train the ML model), the UE requests from the network (with a De-Activate ML model Request message) to be switched to a different operation; optionally, the UE may additionally send an ML State Indication message to the network to inform the network about updating its Default Behavior for the problem pm; this can be the case when the UE detects it is not capable for full ML processing for the current state; the network acknowledges the request in the De-Activate ML model Response message].
Regarding Claim 55,
further comprising determining autonomously to deactivate the at least one ML-based feature and activate the at least one fallback feature [WO8037; p. 14 / lines 14-17, 19-26; at some point in time, UE detects a State Change that affects its ability to execute and/or train the ML model; in this situation, the UE can declare to the network it is not able for full ML processing (for instance using Option a or Option b), and UE autonomously falls back to Default Behavior; in the example of Fig. 5, activation of an ML model and detecting the UE state change are the same as in the example of Fig. 4. However, differently than in Fig. 4, if the UE detects its state change (inability to execute and/or train the ML model), the UE requests from the network (with a De-Activate ML model Request message) to be switched to a different operation; optionally, the UE may additionally send an ML State Indication message to the network to inform the network about updating its Default Behavior for the problem pm; this can be the case when the UE detects it is not capable for full ML processing for the current state; the network acknowledges the request in the De-Activate ML model Response message].
Regarding Claim 56, which recites a method performed by a network node for configuring a wireless device for fallback operation of a machine learning (ML) model having the same claim limitations as those in claim 44 above, the same rationale of rejection as presented in claim 44 is applicable.
Regarding Claims 57-66, which recites a network node having the same claim limitations as those in claims 45-54 above, the same rationale of rejection as presented in claims 45-54 is applicable.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Ali et al. (WO 2024/110160) [para. 0062].
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SAAD A WAQAS whose telephone number is (571)270-5642. The examiner can normally be reached 8:30 - 5:00 PM.
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, Marcus Smith can be reached at (571) 270-1096. 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.
SAAD A. WAQAS
Primary Examiner
Art Unit 2468
/Saad A. Waqas/Primary Examiner, Art Unit 2468
Prosecution Insights