DETAILED ACTION
This action is responsive to communications filed 17 October 2024.
Claims 4, 6, 8, 13, 16, 18, 22 and 27-32 have been cancelled.
Claims 1-3, 5, 7, 9-12, 14-15, 17, 19-21, 23-26 and 33 are subject to examination.
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
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 17 October 2024. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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 33 rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because the claim is directed to a signal per se. The claim recites “computer-readable storage medium”; however, this does not prevent the claim from reciting transitory signals. The claim merely denote a medium that is capable of being executed by a computer to perform operations; however, the medium is not defined as being directed solely to non-transitory signals, and therefore may be read upon transitory signals such that the claims are directed to a propagating electrical or electromagnetic signal or carrier wave. The claims do not impose the program as being embodied on any physical component, such as on a computer program product and/or system/machine for storing instructions therefore for use so as to execute functionality. As such, the claims are directed towards any computer storage medium with instructions, which may be considered as transitory signals and therefore it is a signal per se.
An examination of the specification revealed no identification of a “computer-readable storage medium” being directed solely to non-transitory computer readable mediums or excluding transitory computer readable mediums.
For example, upon examination of the specification it denotes:
[0077] “… such a storage medium may include, but is not limited to, a floppy disk, optical disk, magneto-optical disk, memory card, memory stick, and the like.” but does not define the term from excluding transitory media or including only non-transitory media. The claims are not patent eligible.
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) 1-3, 5, 7, 11, 14-15, 21, 23-25 and 33 is/are rejected under 35 U.S.C. 103 as being unpatentable over Butt et al. (US-20240037450-A1) hereinafter Butt in view of Hu et al. (US-20250209383-A1) hereinafter Hu.
Regarding claim 1, Butt discloses:
An electronic device used on a network device side in a radio communication system ([0001] relate to mobile or wireless telecommunications systems … 5G radio access technology [0031] gNB, see [FIG. 1] device on network side), comprising a processing circuitry ([0070-0072] apparatus 10 … may be a node or element in a communications network … processor 12 for processing information) configured to:
determine at least one first-level federated learning (FL) server entity and a plurality of FL participant entities corresponding to each first-level FL server entity ([0041] by using a ML classification technique, the MTH may form clusters of DTHs/UEs based on correlation of local training data, see [0046] UEs (e.g., DTH) … gNB (e.g., MTH), see [FIG. 4] (i.e. a cluster of UEs (i.e. plurality of FL participant entities) for a gNB (i.e. first-level federated learning (FL) server entity))), wherein one first-level FL server entity and its corresponding plurality of FL participant entities collectively form a group ([0041] cluster, see above in [0046] and [FIG. 4], e.g. grouped UEs for a gNB); and
transmit information of the formed group to the at least one first-level FL server entity ([0054] MTH may obtain information for UEs in a specific cluster), to enable federated learning to be performed within each group ([0028] cluster based host selection in asynchronous federated learning model collection [0067] clustering one or more network elements … selecting at least one network element from the cluster for model training … receiving … updated local model parameters (i.e. from participant in the group is federated learning within the group)).
Butt does not explicitly disclose:
the at least one first-level FL server entity can function as an FL participant of a second-level FL server entity comprised in the electronic device to perform federated learning with the second-level FL server entity; and
transmit information of the formed group to the FL participant entities corresponding to each first-level FL server entity,
However, Hu discloses:
the at least one first-level FL server entity can function as an FL participant of a second-level FL server entity comprised in the electronic device to perform federated learning with the second-level FL server entity ([0029] federated node (i.e. first-level FL server) may serve as one of many federated nodes to the central server (i.e. second-level FL server) … at the same time may serve as a sub-level central server to its sub-level federated nodes (i.e. participant entities) [FIG. 3] e.g. central server 311 with nodes 312, 313, 314; and e.g. node 314 serving as a sub-level central server to nodes 323, 324); and
transmit information of the formed group to the FL participant entities corresponding to each first-level FL server entity ([0029] federated node may host another knowledge graph system … generate a federated knowledge graph to coordinate that federated node and other sub-level federated nodes in its sub-network),
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have the at least one first-level FL server entity function as an FL participant of a second-level FL server entity and transmitted information of the formed group to the FL participant entities corresponding to each first-level FL server entity. One of ordinary skill in the art would have been motivated to do so to have a number of knowledge graph systems hosted on one or more local computing systems and the central server to coordinate the server and the federated nodes, and the local computing system coordinate that federated node and the other sub-level federated nodes to optimize the federated learning processes at different sub-network levels (Hu, [0029]).
Regarding claim 2, Butt-Hu disclose:
The electronic device according to claim 1, wherein the processing circuitry, set forth above, is further configured to:
Butt discloses:
receive, from terminal devices within a coverage area of the electronic device ([0045] DTH in a coverage area), one or more of processing capability, machine learning capability, geographic location ([0045] geographical proximity, see [0040] physical proximity may be used to evaluate candidates to be in the same cluster (i.e. requires to obtain geographic location information to determine)), radio channel quality and movement trajectory information of the terminal devices;
wherein each manager device comprises a first-level FL server entity ([0041] by using a ML classification technique, the MTH may form clusters of DTHs/UEs based on correlation of local training data, see [0046] UEs (e.g., DTH) … gNB (e.g., MTH), see [FIG. 4] (i.e. a cluster of UEs (i.e. plurality of FL participant entities) for a gNB (i.e. first-level federated learning (FL) server entity, MTH))); and
for each manager device, determine terminal devices within a predetermined distance from the manager device as comprising FL participant entities corresponding to the first-level FL server entity comprised in the manager device ([0045] geographical proximity, see [0040] physical proximity may be used to evaluate candidates to be in the same cluster).
Butt does not explicitly disclose:
according to the received one or more of the processing capacity, the machine learning capability, the geographic location, the radio channel quality and the movement trajectory information of the terminal devices, select at least one manager device among the terminal devices,
However, Hu discloses:
according to the received one or more of the processing capacity, the machine learning capability ([0048] enable graph learning to identify an effective set of nodes, models and parameters or parameter groups for global training, see [0028] intelligent logic module … learn new knowledge through graph learning based on … ML models, see also [0005] determine that particular ML models … should be distributed to particulate locate sites because these sites have the data needed to train these ML models), the geographic location, the radio channel quality and the movement trajectory information of the terminal devices, select at least one manager device among the terminal devices ([0028] identify new relationships and discover hidden relationships between the entities as represented by corresponding graph nodes in the knowledge graph based on inferences determined through graph learning [0029] used to coordinate the server and the federated nodes that directly connected to the server … sub-level federated nodes (i.e. identifying relationships to coordinate servers is to select, based on relationship, which devices are on which level for federated learning, see [FIG. 3])),
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have selected at least one manager device among the terminal devices according to the received one or more of the machine learning capability. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 3, Butt-Hu disclose:
The electronic device according to claim 2, set forth above,
Butt does not explicitly disclose:
wherein at least one manager device further comprises an FL participant entity, and the first-level FL server entity and the FL participant entity comprised in the manager device are in a same group; or
wherein terminal devices outside the predetermined distance from each manager device are determined as comprising FL entities which function as FL participants to perform federated learning with the second-level server entity together with the first-level FL server entity.
However, Hu discloses:
wherein at least one manager device further comprises an FL participant entity ([0029] federated node (i.e. first-level FL server) may serve as one of many federated nodes to the central server (i.e. second-level FL server) … at the same time may serve as a sub-level central server to its sub-level federated nodes (i.e. participant entities) [FIG. 3] e.g. central server 311 with nodes 312, 313, 314 (i.e. participant to node 311); and e.g. node 314 serving as a sub-level central server to nodes 323, 324), and the first-level FL server entity and the FL participant entity comprised in the manager device are in a same group ([0029] federated node (i.e. first-level FL server) may serve as one of many federated nodes to the central server (i.e. second-level FL server) … at the same time may serve as a sub-level central server to its sub-level federated nodes (i.e. participant entities where node 314 is the first-level FL server entity) [FIG. 3] e.g. central server 311 with nodes 312, 313, 314; and e.g. node 314 serving as a sub-level central server to nodes 323, 324 (i.e. same group as 323/324 belong to 314 and not 312/313));
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have at least one manager device further comprise an FL participant entity and the first-level FL server entity and FL participant entity in the manager device are in a same group. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 5, Butt-Hu disclose:
The electronic device according to claim 1, set forth above,
Butt discloses:
wherein in a procedure of performing federated learning within a group ([0030] federated learning … model are trained at the different distributed hosts [0031] UE1, UE2, UE3 (DTH) … central meta training host (MTH, e.g. gNB)), the FL participant entities of the group upload local-model related information to the first-level FL server entity of the group using direct communication through links between terminal devices and the electronic device ([0031] model may be transmitted to a gNB, see [FIG. 1] e.g. direct links from the UEs to the gNB), so that the first-level FL server entity updates local-models of the FL participant entities of the group by aggregating the received local-model related information ([0030] combines parameters of all the distributed models to generate a main model [0031] aggregates this model and sends it back to the UEs …. fed back to the DTHs for further iterations (i.e. updating the FL participants)), and
wherein the local-model related information is an output result calculated by the FL participant entity according to common data transmitted by the first-level FL server entity based on the local-model ([0048] there may be an aggregated model at the MTH, and the local models of the DTH may have been updated with the aggregated model. Subsequent to the update with the aggregated model, the DTH in a cluster may be updated matured local trained models (i.e. output, e.g. based on data received from MTH, common data as all UE1s receive the same aggregated data and output their own personal updated matured data, see [FIGs. 1-2])).
Regarding claim 7, Butt-Hu disclose:
The electronic device according to claim 1, wherein in a procedure of performing federated learning by the at least one first-level FL server entity and the second-level FL server entity, set forth above, the processing circuitry, set forth above, is further configured to:
Butt does not explicitly disclose:
receive, from each first-level FL server entity of the at least one first-level FL server entity, local global-model related information of the first-level FL server entity, wherein a local global-model of the first-level FL server entity is an aggregation result of local-models of FL participant entities of the group where the first-level FL server entity is; and
update the local global-model of the first-level FL server entity by aggregating the received local global-model related information,
wherein the local global-model related information is an output result calculated by the first-level FL server entity according to common data transmitted by the second-level FL server entity based on the local global-model.
However, Hu discloses:
receive, from each first-level FL server entity of the at least one first-level FL server entity ([0029] federated node (i.e. first-level FL server) may serve as one of many federated nodes to the central server (i.e. second-level FL server) … at the same time may serve as a sub-level central server to its sub-level federated nodes (i.e. participant entities) [FIG. 3] e.g. central server 311 with nodes 312, 313, 314; and e.g. node 314 serving as a sub-level central server to nodes 323, 324), local global-model related information of the first-level FL server entity ([0017] distributed ML models may be separately trained on respective local sites based on corresponding local data … sent back to the central server to be aggregated into the global model, see [FIG. 3] e.g. node 314), wherein a local global-model of the first-level FL server entity is an aggregation result of local-models of FL participant entities of the group where the first-level FL server entity is ([0017] distributed ML models may be separately trained on respective local sites based on corresponding local data … sent back to the central server to be aggregated into the global model, see [FIG. 3] e.g. node 314 with sub-graph of node 323 and 324 to which 314 performs as a sub-central server); and
update the local global-model of the first-level FL server entity by aggregating the received local global-model related information ([0017] send back to the central server to be aggregated into the global model [0020] transmitting the current global model to participating global sites … each round (i.e. updates, iterations, e.g. of aggregated information fed back to be re-iterated)),
wherein the local global-model related information is an output result calculated by the first-level FL server entity according to common data transmitted by the second-level FL server entity based on the local global-model ([0017] central server (i.e. second-level FL server, see [FIG. 3] e.g. node 311) to distribute ML models to a number of local sites (i.e. common data) distributed ML models may be separately trained on respective local sites based on corresponding local data, see [FIG. 3] e.g. information fed back from Nodes 312, 313, 314 to be aggregated by Node 311 and fed back to the corresponding nodes).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have received from each first-level FL server, local global-model related information of the first-level FL server entity to update the local global-model of the first-level FL server by aggregating the received local global-model related information that is an output result calculated by the first-level FL server entity according to common data transmitted by the second-level FL server entity based on the local global-model. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 11, Butt-Hu disclose:
The electronic device according to claim 7, wherein the processing circuitry, set forth above, is further configured to:
Butt does not explicitly disclose:
obtain a global-model of the second-level FL server entity by aggregating the received local global-model related information; and
transmit the global-model to the plurality of FL participant entities corresponding to each first-level FL server entity.
However, Hu discloses:
obtain a global-model of the second-level FL server entity by aggregating the received local global-model related information ([0020] transmitting the current global model to participating local sites, training local models on these local sites to produce a set of potential model updates at each locate site, and aggregating and processing these local updates into a single global update and applying it to the global model); and
transmit the global-model to the plurality of FL participant entities corresponding to each first-level FL server entity ([0020] transmitting the current global model to participating local sites, training local models on these local sites to produce a set of potential model updates at each locate site, and aggregating and processing these local updates into a single global update and applying it to the global model, see [FIG. 3] sub-graph of node 314 with information that is fed to Nodes 323 and 324 for updating).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have obtained a global-model of the second-level FL server by aggregating the received local global-model related information to transmit the global-model to the plurality of FL participant entities corresponding to each first-level FL server entity. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 14, Butt discloses:
An electronic device used on a user equipment side in a radio communication system ([0001] relate to mobile or wireless telecommunications systems … 5G radio access technology [0031] UE1, UE2, UE3, see [FIG. 1] device on user equipment side), comprising a processing circuitry ([0070-0072] apparatus 10 … may be a node or element in a communications network … processor 12 for processing information) configured to:
receive, from an electronic device on a network device side ([0045] DTH in a coverage area, see [FIG. 1] e.g. UE as above), information of a group where the electronic device on the user equipment side is ([0045] geographical proximity, see [0040] physical proximity may be used to evaluate candidates to be in the same cluster (i.e. requires to obtain geographic location information to determine)), wherein the group comprises one first-level FL server entity and its corresponding plurality of FL participant entities ([0041] by using a ML classification technique, the MTH may form clusters of DTHs/UEs based on correlation of local training data, see [0046] UEs (e.g., DTH) … gNB (e.g., MTH), see [FIG. 4] (i.e. a cluster of UEs (i.e. plurality of FL participant entities) for a gNB (i.e. first-level federated learning (FL) server entity))); and
perform federated learning within the group based on the information of the group ([0028] cluster based host selection in asynchronous federated learning model collection [0067] clustering one or more network elements … selecting at least one network element from the cluster for model training … receiving … updated local model parameters (i.e. from participant in the group is federated learning within the group)).
Butt does not explicitly disclose:
at least one first-level FL server entity determined by the electronic device on the network device side can function as an FL participant of a second-level FL server entity comprised in the electronic device on the network device side to perform federated learning with the second-level FL server entity.
However, Hu discloses:
at least one first-level FL server entity determined by the electronic device on the network device side can function as an FL participant of a second-level FL server entity comprised in the electronic device on the network device side to perform federated learning with the second-level FL server entity ([0029] federated node (i.e. first-level FL server) may serve as one of many federated nodes to the central server (i.e. second-level FL server) … at the same time may serve as a sub-level central server to its sub-level federated nodes (i.e. participant entities) [FIG. 3] e.g. central server 311 with nodes 312, 313, 314; and e.g. node 314 serving as a sub-level central server to nodes 323, 324).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have the at least one first-level FL server entity function as an FL participant of a second-level FL server entity. One of ordinary skill in the art would have been motivated to do so to have a number of knowledge graph systems hosted on one or more local computing systems and the central server to coordinate the server and the federated nodes, and the local computing system coordinate that federated node and the other sub-level federated nodes to optimize the federated learning processes at different sub-network levels (Hu, [0029]).
Regarding claim 15, Butt-Hu disclose:
The electronic device according to claim 14, set forth above,
Butt discloses:
wherein in a case where the electronic device comprises a first-level FL server entity ([0041] by using a ML classification technique, the MTH may form clusters of DTHs/UEs based on correlation of local training data, see [0046] UEs (e.g., DTH) … gNB (e.g., MTH), see [FIG. 4] (i.e. a cluster of UEs (i.e. plurality of FL participant entities) for a gNB (i.e. first-level federated learning (FL) server entity, MTH))), the processing circuitry is configured to:
receive, from FL participant entities of the group where the electronic device is ([0041] by using a ML classification technique, the MTH may form clusters of DTHs/UEs based on correlation of local training data, see [0046] UEs (e.g., DTH) … gNB (e.g., MTH), see [FIG. 4] (i.e. a cluster of UEs (i.e. plurality of FL participant entities) for a gNB (i.e. first-level federated learning (FL) server entity, MTH))), local-model related information ([0031] model may be transmitted to a gNB, see [FIG. 1] e.g. direct links from the UEs to the gNB); and
update local-models of the FL participant entities of the group by aggregating the received local-model related information ([0030] combines parameters of all the distributed models to generate a main model [0031] aggregates this model and sends it back to the UEs …. fed back to the DTHs for further iterations (i.e. updating the FL participants)),
wherein the local-model related information is an output result calculated by the FL participant entity according to common data transmitted by the first-level FL server entity based on the local-model ([0048] there may be an aggregated model at the MTH, and the local models of the DTH may have been updated with the aggregated model. Subsequent to the update with the aggregated model, the DTH in a cluster may be updated matured local trained models (i.e. output, e.g. based on data received from MTH, common data as all UE1s receive the same aggregated data and output their own personal updated matured data, see [FIGs. 1-2])).
Regarding claim 21, Butt-Hu disclose:
The electronic device according to claim 14, set forth above,
Butt discloses:
wherein in a case where the electronic device comprises the first-level FL server entity [0041] by using a ML classification technique, the MTH may form clusters of DTHs/UEs based on correlation of local training data, see [0046] UEs (e.g., DTH) … gNB (e.g., MTH), see [FIG. 4] (i.e. a cluster of UEs (i.e. plurality of FL participant entities) for a gNB (i.e. first-level federated learning (FL) server entity, MTH))), the processing circuitry is configured to:
Butt does not explicitly disclose:
transmit local global-model related information to the second-level FL server entity, so that the second-level FL server entity updates a local global-model of the first-level FL server entity by aggregating the received local global-model related information, wherein the local global-model of the first-level FL server entity is an aggregation result of local-models of the FL participant entities of the group where the first-level FL server entity is,
wherein the local global-model related information is an output result calculated by the first-level FL server entity according to common data transmitted by the second-level FL server entity based on the local global-model.
However, Hu discloses:
transmit local global-model related information to the second-level FL server entity ([0029] federated node (i.e. first-level FL server) may serve as one of many federated nodes to the central server (i.e. second-level FL server) … at the same time may serve as a sub-level central server to its sub-level federated nodes (i.e. participant entities) [FIG. 3] e.g. central server 311 with nodes 312, 313, 314; and e.g. node 314 serving as a sub-level central server to nodes 323, 324), so that the second-level FL server entity updates a local global-model of the first-level FL server entity by aggregating the received local global-model related information ([0017] send back to the central server to be aggregated into the global model [0020] transmitting the current global model to participating global sites … each round (i.e. updates, iterations, e.g. of aggregated information fed back to be re-iterated)), wherein the local global-model of the first-level FL server entity is an aggregation result of local-models of the FL participant entities of the group where the first-level FL server entity is ([0017] distributed ML models may be separately trained on respective local sites based on corresponding local data … sent back to the central server to be aggregated into the global model, see [FIG. 3] e.g. node 314 with sub-graph of node 323 and 324 to which 314 performs as a sub-central server),
wherein the local global-model related information is an output result calculated by the first-level FL server entity according to common data transmitted by the second-level FL server entity based on the local global-model ([0017] central server (i.e. second-level FL server, see [FIG. 3] e.g. node 311) to distribute ML models to a number of local sites (i.e. common data) distributed ML models may be separately trained on respective local sites based on corresponding local data, see [FIG. 3] e.g. information fed back from Nodes 312, 313, 314 to be aggregated by Node 311 and fed back to the corresponding nodes).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have received from each first-level FL server, local global-model related information of the first-level FL server entity to update the local global-model of the first-level FL server by aggregating the received local global-model related information that is an output result calculated by the first-level FL server entity according to common data transmitted by the second-level FL server entity based on the local global-model. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 23, Butt-Hu disclose:
The electronic device according to claim 21, wherein the processing circuitry is further configured to:
Butt does not explicitly disclose:
exchange the local global-model related information with another first-level FL server entity.
However, Hu discloses:
exchange the local global-model related information with another first-level FL server entity ([0017] send back to the central server to be aggregated into the global model [0020] transmitting the current global model to participating global sites … each round (i.e. updates, iterations, e.g. of aggregated information fed back to be re-iterated)).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have exchanged the local global-model related information with another first-level FL server entity. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 24, Butt-Hu disclose:
The electronic device according to claim 14, set forth above,
Butt discloses:
wherein in a case where the electronic device comprises an FL participant entity ([0031] model may be transmitted to a gNB, see [FIG. 1] e.g. direct links from the UEs to the gNB), the processing circuitry is further configured to:
transmit local-model related information to the first-level FL server entity of the group where the electronic device is ([0031] model may be transmitted to a gNB, see [FIG. 1] e.g. direct links from the UEs to the gNB), so that the first-level FL server entity updates local-models of FL participant entities of the group by aggregating the local-model related information received from the FL participant entities of the group ([0030] combines parameters of all the distributed models to generate a main model [0031] aggregates this model and sends it back to the UEs …. fed back to the DTHs for further iterations (i.e. updating the FL participants)).
Regarding claim 25, Butt-Hu disclose:
The electronic device according to claim 24, wherein the processing circuitry, set forth above, is further configured to:
Butt discloses:
receive common data from the first-level FL server entity ([0048] there may be an aggregated model at the MTH, and the local models of the DTH may have been updated with the aggregated model. Subsequent to the update with the aggregated model, the DTH in a cluster may be updated matured local trained models (i.e. output, e.g. based on data received from MTH, common data as all UE1s receive the same aggregated data and output their own personal updated matured data, see [FIGs. 1-2]));
calculate an output result based on the local-model according to the common data ([0048] there may be an aggregated model at the MTH, and the local models of the DTH may have been updated with the aggregated model. Subsequent to the update with the aggregated model, the DTH in a cluster may be updated matured local trained models (i.e. output, e.g. based on data received from MTH, common data as all UE1s receive the same aggregated data and output their own personal updated matured data, see [FIGs. 1-2])); and
Butt does not explicitly disclose:
transmit the output result to another FL participant entity within the same group.
However, Hu discloses:
transmit the output result to another FL participant entity within the same group ([0033] relationships for sharing models and parameters through the network connections may be represented by the edges, see [FIG. 3] e.g. between node 313 and 314 that are participants of node 311).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have transmitted the output result to another FL participant entity within the same group. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Regarding claim 33, it does not further define nor teach over the limitations of claim 1 or claim 14, therefore, claim 33 is rejected for at least the same reasons set forth above as in claim 1 or claim 14.
Claim(s) 9-10, 17, 19 and 26 is/are rejected under 35 U.S.C. 103 as being unpatentable over Butt et al. (US-20240037450-A1) hereinafter Butt in view of Hu et al. (US-20250209383-A1) hereinafter Hu further in view of Butt et al. (US-20240152768-A1) hereinafter Butt(2).
Regarding claim 9, Butt-Hu disclose:
The electronic device according to claim 7, wherein the processing circuitry, set forth above, is further configured to:
Butt-Hu do not explicitly disclose:
receive, from each first-level FL server entity, two or more of a first quantity related to local global-model prediction accuracy, a second quantity related to channel quality, and a third quantity related to historical and/or future trajectory, of the terminal device where the first-level FL server entity is; and
determine a weight corresponding to the first-level FL server entity based on the two or more of the first quantity, the second quantity and the third quantity,
wherein the local global-model of the first-level FL server entity is updated by weighted aggregation of the output result received from the first-level FL server entity using the weight corresponding to the first-level FL server entity.
However, Butt(2) discloses:
receive, from each first-level FL server entity ([0246] additional gNB information), two or more of a first quantity related to local global-model prediction accuracy, a second quantity related to channel quality, and a third quantity related to historical and/or future trajectory, of the terminal device where the first-level FL server entity is ([0253-0256] additional UE-related non-radio information … UE location … movement vector/acceleration (i.e. trajectory) … UE historical FL model accuracy (i.e. prediction accuracy)); and
determine a weight corresponding to the first-level FL server entity based on the two or more of the first quantity, the second quantity and the third quantity ([0139-0143] competent entity uses a weighting for calculating a selection index of each DTH … Computational_Power … Battery Energy, see [0253-0265] UE performance (e.g., computation power, communication power and/or energy (budget); i.e. weighted based on the information)),
wherein the local global-model of the first-level FL server entity is updated by weighted aggregation of the output result received from the first-level FL server entity using the weight corresponding to the first-level FL server entity ([0144] priority of the existing/present cluster head naturally goes down in the next round, as Battery_Energy will go down for the cluster head after it performs joint model update/averaging for the whole cluster (i.e. aggregation)).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Butt(2) to have received from each first-level FL server entity, two or more of quantities related to prediction accuracy, channel quality, historical/future trajectory, to determine a weight to update a local global-model by a weighted aggregation. One of ordinary skill in the art would have been motivated to do so to maintain some kind of fairness among the cluster heads, as the role of the cluster head can be switched based on objective criteria (Butt(2), [0145]).
Regarding claim 10, Butt-Hu-Butt(2) disclose:
The electronic device according to claim 9, set forth above,
Butt-Hu do not explicitly disclose:
wherein the weight is calculated as a linear sum of at least two of the first quantity, the second quantity and the third quantity.
However, Butt(2) discloses:
wherein the weight is calculated as a linear sum of at least two of the first quantity, the second quantity and the third quantity ([0139-0143] Head_index=w1*Computational_Power+w2*Battery_Energy).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Butt(2) to have the weight calculated as a linear sum of at least two of the quantities. One of ordinary skill in the art would have been motivated to do so to maintain some kind of fairness among the cluster heads, as the role of the cluster head can be switched based on objective criteria (Butt(2), [0145]).
Regarding claim 17, Butt-Hu disclose:
The electronic device according to claim 15, wherein the processing circuitry, set forth above, is further configured to:
Butt-Hu do not explicitly disclose:
receive, from each FL participant entity in the group where the electronic device is, two or more of a first quantity related to local-model prediction accuracy, a second quantity related to channel quality, and a third quantity related to historical and/or future trajectory, of the terminal device where the FL participant entity is, and
determine a weight corresponding to the FL participant entity based on the two or more of the first quantity, the second quantity, and the third quantity,
wherein the local-model of the FL participant entity of the group is updated by weighted aggregation of the output result received from the FL participant entity using the weight corresponding to the FL participant entity;
or
receive, from the electronic device on the network device side, a weight corresponding to each FL participant entity of the group, the weight being determined by the electronic device on the network device side according to any two or more of a first quantity related to local-model prediction accuracy, a second quantity related to channel quality, and a third quantity related to historical and/or future trajectory, of the terminal device where the FL participant entity is, that are received from each FL participant entity of the group,
wherein the local-model of the FL participant entity of the group is updated by weighted aggregation of the output result received from the FL participant entity using the weight corresponding to the FL participant entity.
However, Butt(2) discloses:
receive, from the electronic device on the network device side ([0246] additional gNB (i.e. network device side) information), a weight corresponding to each FL participant entity of the group ([0139-0143] competent entity uses a weighting for calculating a selection index of each DTH … Computational_Power … Battery Energy, see [0253-0265] UE performance (e.g., computation power, communication power and/or energy (budget); i.e. weighted based on the information)), the weight being determined by the electronic device on the network device side ([0139-0143] competent entity uses a weighting for calculating a selection index of each DTH … Computational_Power … Battery Energy, see [0253-0265] UE performance (e.g., computation power, communication power and/or energy (budget); i.e. weighted based on the information)) according to any two or more of a first quantity related to local-model prediction accuracy, a second quantity related to channel quality, and a third quantity related to historical and/or future trajectory, of the terminal device where the FL participant entity is ([0253-0256] additional UE-related non-radio information … UE location … movement vector/acceleration (i.e. trajectory) … UE historical FL model accuracy (i.e. prediction accuracy)), that are received from each FL participant entity of the group ([FIG. 1] multiple TH’s, e.g. UEs),
wherein the local-model of the FL participant entity of the group is updated by weighted aggregation of the output result received from the FL participant entity using the weight corresponding to the FL participant entity ([0144] priority of the existing/present cluster head naturally goes down in the next round, as Battery_Energy will go down for the cluster head after it performs joint model update/averaging for the whole cluster (i.e. aggregation)).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Butt(2) to have received from each FL participant entity, two or more of quantities related to prediction accuracy, channel quality, historical/future trajectory, to determine a weight to update a local-model by a weighted aggregation. One of ordinary skill in the art would have been motivated to do so to maintain some kind of fairness among the cluster heads, as the role of the cluster head can be switched based on objective criteria (Butt(2), [0145]).
Regarding claim 19, Butt-Hu-Butt(2) disclose:
The electronic device according to claim 17, set forth above,
Butt-Hu do not explicitly disclose:
wherein the weight is calculated as a linear sum of at least two of the first quantity, the second quantity and the third quantity.
However, Butt(2) discloses:
wherein the weight is calculated as a linear sum of at least two of the first quantity, the second quantity and the third quantity ([0139-0143] Head_index=w1*Computational_Power+w2*Battery_Energy).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Butt(2) to have the weight calculated as a linear sum of at least two of the quantities. One of ordinary skill in the art would have been motivated to do so to maintain some kind of fairness among the cluster heads, as the role of the cluster head can be switched based on objective criteria (Butt(2), [0145]).
Regarding claim 26, Butt-Hu disclose:
The electronic device according to claim 25, wherein the processing circuitry, set forth above, is further configured to:
Butt-Hu do not explicitly disclose:
transmit a request message to the other FL participant entity through synchronization channel information (SCI), and
transmit the output result to the other FL participant entity by carrying information for demodulating and decoding a physical side link control channel (PSSCH) by the SCI and by carrying the output result by the PSSCH;
or
transmit the output result to an FL participant entity within a different group through a PC5 link;
or
transmit the output result to the first-level FL server entity in the same group and/or a different group in a manner of D2D.
However, Butt(2) discloses:
transmit the output result to the first-level FL server entity in the same group in a manner of D2D ([0090] local data set for local model training is communicated to the cluster head, preferably via device-to-device (D2D) and/or sidelink communication).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Butt(2) to have transmitted the output result to the first-level FL server entity in the same group in a manner of D2D. One of ordinary skill in the art would have been motivated to do so to enable a one-to-one communication (Butt(2), [0083]).
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Butt et al. (US-20240037450-A1) hereinafter Butt in view of Hu et al. (US-20250209383-A1) hereinafter Hu further in view of Hu et al. (US-20230106985-A1) hereinafter Hu(2).
Regarding claim 12, Butt-Hu disclose:
The electronic device according to claim 1, set forth above,
Butt-Hu do not explicitly disclose:
wherein the information of the formed group comprises one or more of: an identifier ID of a terminal device where the first-level FL server entity in the group is and a group ID of the group; or
wherein the information of the formed group comprises one or more of: an identifier ID of a terminal device where the first-level FL server entity in the group is and a group ID of the group, and
the information of the formed group further comprises: a group ID of a group in an adjacent geographic location.
However, Hu(2) discloses:
wherein the information of the formed group comprises one or more of: an identifier ID of a terminal device where the first-level FL server entity in the group is and a group ID of the group ([0119] Grand Master node may store, for each distributed node, an identifier and address of the distributed node … also store a federation group ID);
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Hu(2) to have the information of the formed group comprise an identifier of a terminal device where the first-level FL server entity in the group is and a group ID of the group. One of ordinary skill in the art would have been motivated to do so to use identifiers for communication and storage purposes (Hu(2), [0119]).
Claim(s) 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Butt et al. (US-20240037450-A1) hereinafter Butt in view of Hu et al. (US-20250209383-A1) hereinafter Hu further in view of Das Gupta et al. (US-11323514-B2) hereinafter Gupta.
Regarding claim 20, Butt-Hu disclose:
The electronic device according to claim 15, wherein the processing circuitry, set forth above, is further configured to:
Butt does not explicitly disclose:
obtain a local global-model by aggregating the received local-model related information, wherein the local global-model of the first-level FL server entity is an aggregation result of the local-models of the FL participant entities of the group where the first-level FL server entity is; and
according to a request from an FL participant entity of another group, transmit the local global-model to the FL participant entity.
However, Hu discloses:
obtain a local global-model by aggregating the received local-model related information ([0020] transmitting the current global model to participating local sites, training local models on these local sites to produce a set of potential model updates at each locate site, and aggregating and processing these local updates into a single global update and applying it to the global model), wherein the local global-model of the first-level FL server entity is an aggregation result of the local-models of the FL participant entities of the group where the first-level FL server entity is ([0020] transmitting the current global model to participating local sites, training local models on these local sites to produce a set of potential model updates at each locate site, and aggregating and processing these local updates into a single global update and applying it to the global model); and
transmit the local global-model to the FL participant entity ([0020] transmitting the current global model to participating local sites, training local models on these local sites to produce a set of potential model updates at each locate site, and aggregating and processing these local updates into a single global update and applying it to the global model, see [FIG. 3] sub-graph of node 314 with information that is fed to Nodes 323 and 324 for updating).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt in view of Hu to have obtained a global-model of the second-level FL server by aggregating the received local global-model related information to transmit the global-model to the plurality of FL participant entities corresponding to each first-level FL server entity. One of ordinary skill in the art would have been motivated to do so to identify an effective set of nodes, models and parameters or parameter groups for global training (Hu, [0048]).
Butt-Hu do not explicitly disclose:
according to a request from an FL participant entity of another group, transmit the local global-model to the FL participant entity.
However, Gupta discloses:
according to a request from an FL participant entity of another group ([8:47-51] body of user requests may be used to identify and/or evaluate the video data which have been requested), transmit the local global-model to the FL participant entity ([8:47-51] global model is tuned to in accordance to the user’s requests … updates to the model include changes that allow the model to identify data relevant to this request to be identified and pushed to the edge servers).
It would have been obvious to one of ordinary skill in the pertinent art before the effective filing date of the claimed invention to modify the invention of Butt-Hu in view of Gupta to transmit the local global-model according to a request from a FL participant entity of another group. One of ordinary skill in the art would have been motivated to do so to include changes that allow the model to identify data relevant to the request to be identified and pushed to the edge servers (Gupta, [8:47-51]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Kopp et al. (US-11373115-B2) Asynchronous Parameter Aggregation For Machine Learning;
Zhou et al. (US-12093837-B2) Building A Federated Learning Framework;
Shaloudegi et al. (US-20210365841-A1) METHODS AND APPARATUSES FOR FEDERATED LEARNING;
Arcot Desai et al. (US-12333443-B2) Systems And Methods For Using Federated Learning For Training Centralized Seizure Detection And Prediction Models On Decentralized Datasets;
Mayyuri et al. (US-11777812-B2) Zone-based Federated Learning;
Guo (US-20230394365-A1) FEDERATED LEARNING PARTICIPANT SELECTION METHOD AND APPARATUS, AND DEVICE AND STORAGE MEDIUM;
Wang et al. (US-20230145177-A1) FEDERATED LEARNING METHOD AND FEDERATED LEARNING SYSTEM BASED ON MEDIATION PROCESS;
Wouhaybi (US-20220222583-A1) APPARATUS, ARTICLES OF MANUFACTURE, AND METHODS FOR CLUSTERED FEDERATED LEARNING USING CONTEXT DATA.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Alex Tran whose telephone number is (571)272-8173. The examiner can normally be reached Monday-Friday 10AM-6PM ET.
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, Kamal Divecha can be reached at (571)272-5863. 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.
/Alex Tran/Primary Examiner, Art Unit 2453
Prosecution Insights