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 Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-30 are rejected under 35 U.S.C. 103 as being unpatentable over Banerjee et al. (US20230169356A1) in view of Jha et al. (US20200027022A1).
Regarding claim 1, Banerjee teaches an apparatus for wireless communication, comprising: a processor; and memory coupled with the processor, the processor configured to: (Paragraph [0279]: The processor 1104 in the first node 111 may be further configured to transmit or send information to e.g., the second node 112, the third node 113, the another node 114, and/or, any other nodes in the first plurality of nodes 110, any of the devices in the plurality of client computing devices 120, any node in the cloud 115, and/or or another structure in the communications system 100, through a sending port 1107, which may be in communication with the processor 1104, and the memory 1105.)
provide a first message including federated learning (FL) information (Paragraph [0084]: In order to eventually identify the clusters or sets of nodes, that may be useful to process data in order to predict the event, in this Action 301, the first node 111 may receive, respectively from each of the nodes in the first plurality of nodes 110, respective first information. According to a first option, the respective first information may be regarding: an arrival of service requests received over a first period of time by the first plurality of nodes 110 from a plurality of client computing devices 130 of the first plurality of nodes 110. A service request may be for example, a traffic condition service or a weather condition service. According to this option, the first node 111 may detect the pattern of arrival of services requests from clients. The first period of time may be, e.g., during training phase. Also see paragraphs [0085], [0086], [0095], [0221]).)
provide a second message indicating a first network node (Paragraph [0137]: In this Action 403, the second node 112 sends a second indication to the nodes in the first set of nodes 121. The second indication indicates the determined vote.)
and obtain a third message indicating one of the first network node or a second network node as an FL cluster leader and indicating an FL cluster of the apparatus based on the FL information (Paragraph [0145]: In this Action 406, the second node 112 may exchange a third indication with the nodes in the first set of nodes 121. The third indication may indicate that the third node 113 is the leader node 113 for the first set of nodes 121. The exchanging of the third indication may be based on the sent second indication in Action 403.)
Banerjee does not explicitly teach indicating an FL cluster of the apparatus based on the FL information.
However, Jha teaches indicating an FL cluster of the apparatus based on the FL information (Paragraph [0077]: The LC may then send the Participant-Solicitation-Request-Packet-B-1-LC to potential participant nodes in proximity to the LC (e.g., as a broadcast message). Here proximity may include network proximity (e.g., nodes within one, two, or some other defined number of hops from the LC) or geographic proximity (e.g., within a certain radius of the LC, within a city of the LC, etc.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide indicating an FL cluster of the apparatus based on the FL information, as taught by Jha in the system of Banerjee, so that the local coordinator can indicate to the participating nodes that they are part of the cluster (Jha: Paragraph [0077]).
Regarding claim 2, the combination of Banerjee and Jha teaches the apparatus of claim 1 (see rejection for claim 1);
Banerjee further teaches wherein FL information comprises information related to at least one of: a machine learning task of the apparatus (Paragraph [0085]: According to a second option, the respective first information may be regarding data collected for service requests received by a second plurality of nodes 119 in the communications system 100 over a second period of time. The service requests may be of different types, for example weather and traffic, or of the same type, for example, two different measures of pollution.)
or an available computation resource of the apparatus for the machine learning task (Paragraph [0086]: According to a third option, the respective first information may be regarding data collected by the first plurality of nodes 110 over the first period of time, about the service requests received. In this case, the data analysed may be contextually similar data, that is data of the same type.)
Banerjee does not explicitly teach and wherein each of the first network node and the second network node is a road side unit (RSU) or an edge server; and wherein the apparatus is a UE.
However, Jha teaches and wherein each of the first network node and the second network node is a road side unit (RSU) or an edge server (Paragraph [0048]: In the second case, an LC is used to assist the GA. For example, in vehicle to everything (V2X) use cases, there may be RSUs used as LCs. Similarly, in case of smart building (e.g., smart home, smart office, etc.), IoT gateways or the like may be used as LCs.)
and wherein the apparatus is a UE (Paragraph [0046]: Generally, the examples assume that the data used for training is stored at the participant devices, such as mobile phones, tablets, vehicles, home sensors, etc.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide wherein each of the first network node and the second network node is a road side unit (RSU) or an edge server; and wherein the apparatus is a UE, as taught by Jha in the system of Banerjee, so that the local coordinator (RSU) can route information centric networking messages between the participant devices such as UEs and the global aggregator, and help relieve network congestion (Jha: Paragraphs [0022], [0046], [0048]).
Regarding claim 3, the combination of Banerjee and Jha teaches the apparatus of claim 1, wherein the processor is configured to (see rejection for claim 1);
Banerjee does not explicitly teach to provide the first message periodically or in response to an event trigger.
However, Jha teaches to provide the first message periodically or in response to an event trigger (Paragraph [0071]: Devices that satisfying requirements specified by the GA in the Periodic-DML-Participant-Solicitation-Request-Packet-P-1 may send a Periodic-DML-Participant-Response-Data-Packet-P-2 in response to the Periodic-DML-Participant-Solicitation-Request-Packet-P-1. Paragraph [0096]: The LC may maintain this data via updates pushed from the devices or via a periodical discovery broadcast or directed query of the devices. Also see paragraph [0095]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the first message periodically or in response to an event trigger, as taught by Jha in the system of Banerjee, so that the local coordinator can send periodic information to the global coordinator (Jha: Paragraphs: [0071], [0095], [0096]).
Regarding claim 4, the combination of Banerjee and Jha teaches the apparatus of claim 1, wherein the processor is configured to (see rejection for claim 1);
Banerjee further teaches to provide the first message in a broadcast (Paragraph [0084]: In order to eventually identify the clusters or sets of nodes, that may be useful to process data in order to predict the event, in this Action 301, the first node 111 may receive, respectively from each of the nodes in the first plurality of nodes 110, respective first information. Paragraph [0087]: The receiving in this Action 301 may be implemented through a peer-to-peer, or broadcast, protocol.)
Regarding claim 5, the combination of Banerjee and Jha teaches the apparatus of claim 1, wherein the processor is configured to: (see rejection for claim 1);
Banerjee further teaches to obtain an acknowledgment of the FL information from the first network node, wherein the processor is configured to provide the second message in response to the acknowledgment (Paragraph [0126]: In this Action 401 the second node 112 may receive the first indication from the first node 111 comprised in the communications system 100. Paragraph [0135]: In some embodiments, the determining in this Action 402 may be triggered by the receiving of the first indication in Action 401.)
Regarding claim 6, the combination of Banerjee and Jha teaches the apparatus of claim 1, wherein the processor is configured to (see rejection for claim 1);
Banerjee further teaches to obtain the third message indicating the FL cluster leader and the FL cluster of the apparatus in a groupcast from the first network node (Paragraph [0137]: In this Action 403, the second node 112 sends a second indication to the nodes in the first set of nodes 121. The second indication indicates the determined vote. Paragraph [0138]: The sending in this Action 403 may be implemented through a peer-to-peer, or broadcast, protocol.)
Regarding claim 7, the combination of Banerjee and Jha teaches the apparatus of claim 1, wherein the processor is configured to (see rejection for claim 1);
Banerjee further teaches to provide the first message to the second network node, the first message further including a request to join a second FL cluster of the second network node (Paragraph [0140]: Since the other nodes in the first set of nodes 121 may be understood to also cast their respective votes of the leader node, the second node 112 may, in this Action 404, receive a respective vote from each of the other nodes in the first set of nodes 121 for, out of the nodes comprised in the first set of nodes 121, their respective candidate for leader node 113 of the first set of nodes 121. Paragraph [0142]: In this Action 405, the second node 112 may determine the third node 113 to be the leader node for the first set of nodes 121. The determining in this Action 405 may be based on the received respective votes in Action 404.)
Regarding claim 8, the combination of Banerjee and Jha teaches the apparatus of claim 7, wherein the processor is further configured to: (see rejection for claim 7);
Banerjee further teaches to obtain a message indicating the second network node is a candidate to be a second FL cluster leader of the second FL cluster; wherein the processor is configured to provide the first message to the second network node in response to the message (Paragraph [0140]: Since the other nodes in the first set of nodes 121 may be understood to also cast their respective votes of the leader node, the second node 112 may, in this Action 404, receive a respective vote from each of the other nodes in the first set of nodes 121 for, out of the nodes comprised in the first set of nodes 121, their respective candidate for leader node 113 of the first set of nodes 121. Paragraph [0142]: In this Action 405, the second node 112 may determine the third node 113 to be the leader node for the first set of nodes 121. The determining in this Action 405 may be based on the received respective votes in Action 404.)
Regarding claim 9, the combination of Banerjee and Jha teaches the apparatus of claim 8 (see rejection for claim 8);
Banerjee further teaches wherein the second message further includes the FL information; and wherein the processor is configured to provide the second message to the first network node in response to a lack of acknowledgment of the first message from the second network node within a message timeout window; wherein the third message indicates the first network node as the FL cluster leader (Paragraph [0084]: According to a first option, the respective first information may be regarding: an arrival of service requests received over a first period of time by the first plurality of nodes 110 from a plurality of client computing devices 130 of the first plurality of nodes 110. Paragraph [0090]: In this Action 302, the first node 111 may receive, from another node 114 operating in the communications system 100, second information regarding the service requests received over the second period of time. Paragraph [0093]: By receiving the second information in this Action 302, the first node 111 may then be enabled to analyze it in Action 303 to identify one or more clusters of nodes with the purpose of predicting the event. Paragraph [0095]: Once the first node 111 may have gathered the first information and/or the second information, it may be ready to try to identify a cluster of nodes, based on the particular event it may wish to predict. Paragraph [0125]: Once the cluster, that is, the first set of nodes 121 may have been identified, nodes comprised in it, such as the second node 112, may cast a vote to elect the leader node of the newly defined set of nodes 111. Paragraph [0126]: In this Action 401 the second node 112 may receive the first indication from the first node 111 comprised in the communications system 100. As stated earlier, the first indication may indicate that the second node 112 is comprised in the first set of nodes 121. By receiving the first indication, the second node 112 may be enabled to know which potential candidates it may choose from in order to elect the leader node.)
Regarding claim 10, the combination of Banerjee and Jha teaches the apparatus of claim 1, wherein the processor is further configured to: (see rejection for claim 1);
Banerjee does not explicitly teach to obtain, from the FL cluster leader, a machine learning (ML) model configuration including an initial weight; provide, to the FL cluster leader, a ML model information update including an update to the initial weight; and obtain, from the FL cluster leader, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update of a third network node in the FL cluster.
However, Jha teaches to obtain, from the FL cluster leader, a machine learning (ML) model configuration including an initial weight; provide, to the FL cluster leader, a ML model information update including an update to the initial weight (Paragraph [0078]: For Example, the Participant-Solicitation-Request-Packet-B-1-LC may include one of more of the Tentative Round Information, DML Task Details, Communicate-Compute-Storage Requirements, query for Participant-Volume-Information, query for data quality, AoI, TEE requirement for training, or information to access the model training algorithm. Thus, the Participant-Solicitation-Request-Packet-B-1-LC may be thought of as a local version of Participant-Solicitation-Request-Packet-B-1, where locality is defined by the LC. Paragraph [0079]: Interested participant nodes-which may be limited to those that satisfy the requirements specified by the GA-respond with a Participant-Response-Data-Packet-B-2 data packet to the LC (message(s) 715). The Participant-Response-Data-Packet-B-2 identifies the node's intent to participate in the DML training (e.g., a current DML training round). The Participant-Response-Data-Packet-B-2 may include one or more of Availability-Windows for the node, Communicate-Compute-Storage Capabilities of the node; or Data-quality of the node's data that will be used to training the model. In an example, if the device intends to participate in the DML round and does not have the model training algorithm, the device may retrieve the model training algorithm prior to sending the Participant-Response-Data-Packet-B-2.)
and obtain, from the FL cluster leader, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update of a third network node in the FL cluster (Paragraph [0085]: In response to receiving the Model-Download-Data-Packet-B-3, the LC may send a Model-Download-Data-Packet-B-3-LC (message 730) data packet to the selected participant devices. The LC may transmit the Model-Download-Data-Packet-B-3-LC as broadcast (e.g., if several participants are selected) or unicast (e.g., if only one or two participants selected from this LC) message. Paragraph [0086]: At this stage, the participant devices perform model training in accordance with the DML training round criteria (e.g., using the specified model training algorithm over a specified set of data, etc.) to produce an update for the model. The devices may then transmit the produce model update to the LC in a Model-Update-Data-Packet-B-4 data packet (message 735). In an example, the Model-Update-Data-Packet-B-4 data packet is a unicast message to the LC.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to obtain, from the FL cluster leader, a machine learning (ML) model configuration including an initial weight; provide, to the FL cluster leader, a ML model information update including an update to the initial weight; and obtain, from the FL cluster leader, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update of a third network node in the FL cluster, as taught by Jha in the system of Banerjee, so that model updates can be performed by the participant devices, and forwarded to the global aggregator by the local coordinator (Jha: Paragraphs [0078], [0079], [0085], [0086]).
Regarding claim 11, the combination of Banerjee and Jha teaches the apparatus of claim 10 (see rejection for claim 10);
Banerjee does not explicitly teach wherein the aggregated ML model information update further includes a second aggregation of the ML model information update with a third ML model information update of a fourth network node in a second FL cluster.
However, Jha teaches wherein the aggregated ML model information update further includes a second aggregation of the ML model information update with a third ML model information update of a fourth network node in a second FL cluster (Paragraph [0084]: After participant devices are selected, the GA may send a Model-Download-Data-Packet-B-3 (message 725) data packet for this round of DML to each LC. In an example, the Model-Download-Data-Packet-B-3 is unicast to the LCs. In an example, the Model-Download-Data-Packet-B-3 includes a selected participant list if, for example, the participants are selected by the GA or if the GA has overridden one or more of an LC's participant device selections. Paragraph [0086]: At this stage, the participant devices perform model training in accordance with the DML training round criteria (e.g., using the specified model training algorithm over a specified set of data, etc.) to produce an update for the model. The devices may then transmit the produce model update to the LC in a Model-Update-Data-Packet-B-4 data packet (message 735). In an example, the Model-Update-Data-Packet-B-4 data packet is a unicast message to the LC. Paragraph [0087]: Once the aggregated model update is produced, the LC transmits a Model-Update-Data-Packet-B-4-LC (message 740) data packet to the GA. This Model-Update-Data-Packet-B-4-LC includes the aggregated model update. In an example, the Model-Update-Data-Packet-B-4-LC is a unicast message.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide wherein the aggregated ML model information update further includes a second aggregation of the ML model information update with a third ML model information update of a fourth network node in a second FL cluster, as taught by Jha in the system of Banerjee, so that model updates can be performed by the participant devices, and forwarded to the global aggregator by the local coordinator (Jha: Paragraphs [0084], [0086], [0087]).
Regarding claim 12, Banerjee teaches an apparatus for wireless communication, comprising: a processor; and memory coupled with the processor, the processor configured to: obtain a first message including federated learning (FL) information of a first network node; and provide a second message indicating an FL cluster of the first network node based on the FL information, wherein the second message indicates one of the apparatus or a second network node as an FL cluster leader of the FL cluster (see rejection for claim 1);
Banerjee does not explicitly teach indicating an FL cluster of the first network node based on
the FL information.
However, Jha teaches indicating an FL cluster of the first network node based on the FL
information (see rejection for claim 1);
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing
date of the claimed invention to provide indicating an FL cluster of the first network node based on the FL information as taught by Jha in the system of Banerjee, so that the local coordinator can indicate to the participating nodes that they are part of the cluster (Jha: Paragraph [0077]).
Regarding claim 13, the combination of Banerjee and Jha teaches the apparatus of claim 12, wherein the processor is further configured to: (see rejection for claim 12);
Banerjee further teaches to provide an acknowledgment to the first network node in response to the first message; (Paragraph [0126]: In this Action 401 the second node 112 may receive the first indication from the first node 111 comprised in the communications system 100. Paragraph [0135]: In some embodiments, the determining in this Action 402 may be triggered by the receiving of the first indication in Action 401.)
Banerjee does not explicitly teach to obtain a third message from the first network node indicating an identifier of the apparatus or of the second network node in response to the acknowledgment.
However, Jha teaches to obtain a third message from the first network node indicating an identifier of the apparatus or of the second network node in response to the acknowledgment (Paragraph [0080]: In an example, the response packets may also include device attestation evidence according to trustworthiness properties of the device. A device may employ a variety of attestation roots of trust including but not limited to a trusted platform module (TPM), a device identifier composition engine (DICE), software guard extensions (SGX), trusted domain extension (TDX), TrustZone, etc. In an example, the LC or the GC may act as attestation verifiers for provided attestation evidence. In an example, the LC or the GC may have security policies that require verification of attestation evidence as a pre-condition to acting on or forwarding other information contained in the message. In an example, the DML round context (e.g., the Round Start time or the Round End time) may include a session handle or context identifier.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing
date of the claimed invention to obtain a third message from the first network node indicating an identifier of the apparatus or of the second network node in response to the acknowledgment, as taught by Jha in the system of Banerjee, since security policies require verification of the devices before forwarding other information contained in the message (Jha: Paragraph [0080]).
Regarding claim 14, the combination of Banerjee and Jha teaches the apparatus of claim 13, wherein the processor is further configured to: (see rejection for claim 13);
Banerjee further teaches to provide, to a FL parameter network entity, a fourth message including the FL information (Paragraph [0084]: In order to eventually identify the clusters or sets of nodes, that may be useful to process data in order to predict the event, in this Action 301, the first node 111 may receive, respectively from each of the nodes in the first plurality of nodes 110, respective first information. According to a first option, the respective first information may be regarding: an arrival of service requests received over a first period of time by the first plurality of nodes 110 from a plurality of client computing devices 130 of the first plurality of nodes 110. A service request may be for example, a traffic condition service or a weather condition service. According to this option, the first node 111 may detect the pattern of arrival of services requests from clients. The first period of time may be, e.g., during training phase.)
Banerjee does not explicitly teach a second identifier of the first network node in response to the third message indicating the identifier of the apparatus.
However, Jha teaches a second identifier of the first network node in response to the third message indicating the identifier of the apparatus (Paragraph [0080]: In an example, the response packets may also include device attestation evidence according to trustworthiness properties of the device. A device may employ a variety of attestation roots of trust including but not limited to a trusted platform module (TPM), a device identifier composition engine (DICE), software guard extensions (SGX), trusted domain extension (TDX), TrustZone, etc. In an example, the LC or the GC may act as attestation verifiers for provided attestation evidence. In an example, the LC or the GC may have security policies that require verification of attestation evidence as a pre-condition to acting on or forwarding other information contained in the message. In an example, the DML round context (e.g., the Round Start time or the Round End time) may include a session handle or context identifier.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing
date of the claimed invention to provide a second identifier of the first network node in response to the third message indicating the identifier of the apparatus, as taught by Jha in the system of Banerjee, since security policies require verification before forwarding other information contained in the message (Jha: Paragraph [0080]).
Regarding claim 15, the combination of Banerjee and Jha teaches the apparatus of claim 14 (see rejection for claim 14);
Banerjee further teaches wherein the fourth message further includes second FL information of the apparatus (Paragraph [0084]: In order to eventually identify the clusters or sets of nodes, that may be useful to process data in order to predict the event, in this Action 301, the first node 111 may receive, respectively from each of the nodes in the first plurality of nodes 110, respective first information. According to a first option, the respective first information may be regarding: an arrival of service requests received over a first period of time by the first plurality of nodes 110 from a plurality of client computing devices 130 of the first plurality of nodes 110. A service request may be for example, a traffic condition service or a weather condition service. According to this option, the first node 111 may detect the pattern of arrival of services requests from clients. The first period of time may be, e.g., during training phase.)
Regarding claim 16, the combination of Banerjee and Jha teaches the apparatus of claim 13, wherein the processor is further configured to: (see rejection for claim 13);
Banerjee does not explicitly teach to provide, to a FL parameter network entity, a fourth message including a second identifier of the first network node in response to the third message indicating the identifier of the second network node.
However, Jha teaches to provide, to a FL parameter network entity, a fourth message including a second identifier of the first network node in response to the third message indicating the identifier of the second network node (Paragraph [0080]: In an example, the response packets may also include device attestation evidence according to trustworthiness properties of the device. A device may employ a variety of attestation roots of trust including but not limited to a trusted platform module (TPM), a device identifier composition engine (DICE), software guard extensions (SGX), trusted domain extension (TDX), TrustZone, etc. In an example, the LC or the GC may act as attestation verifiers for provided attestation evidence. In an example, the LC or the GC may have security policies that require verification of attestation evidence as a pre-condition to acting on or forwarding other information contained in the message. In an example, the DML round context (e.g., the Round Start time or the Round End time) may include a session handle or context identifier.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing
date of the claimed invention to provide, to a FL parameter network entity, a fourth message including a second identifier of the first network node in response to the third message indicating the identifier of the second network node, as taught by Jha in the system of Banerjee, since security policies require verification before forwarding other information contained in the message (Jha: Paragraph [0080]).
Regarding claim 17, the combination of Banerjee and Jha teaches the apparatus of claim 12, wherein the processor is further configured to: (see rejection for claim 12);
Banerjee further teaches to obtain, from a FL parameter network entity, a third message indicating the FL cluster and indicating the apparatus is the FL cluster leader of the FL cluster; wherein the second message is provided to the first network node in response to the third message (Paragraph [0140]: Since the other nodes in the first set of nodes 121 may be understood to also cast their respective votes of the leader node, the second node 112 may, in this Action 404, receive a respective vote from each of the other nodes in the first set of nodes 121 for, out of the nodes comprised in the first set of nodes 121, their respective candidate for leader node 113 of the first set of nodes 121. Paragraph [0142]: In this Action 405, the second node 112 may determine the third node 113 to be the leader node for the first set of nodes 121. The determining in this Action 405 may be based on the received respective votes in Action 404. Paragraph [0145]: In this Action 406, the second node 112 may exchange a third indication with the nodes in the first set of nodes 121. The third indication may indicate that the third node 113 is the leader node 113 for the first set of nodes 121. The exchanging of the third indication may be based on the sent second indication in Action 403.))
Regarding claim 18, the combination of Banerjee and Jha teaches the apparatus of claim 12, wherein the processor is configured to (see rejection for claim 12);
Banerjee further teaches to groupcast the second message to a plurality of network nodes including the first network node in the FL cluster (Paragraph [0137]: In this Action 403, the second node 112 sends a second indication to the nodes in the first set of nodes 121. The second indication indicates the determined vote. Paragraph [0138]: The sending in this Action 403 may be implemented through a peer-to-peer, or broadcast, protocol.)
Regarding claim 19, the combination of Banerjee and Jha teaches the apparatus of claim 12, wherein the processor is further configured to: (see rejection for claim 12);
Banerjee further teaches to provide a third message to the first network node indicating the apparatus is a candidate to be the FL cluster leader of the FL cluster; wherein the first message is responsive to the third message and indicates a request from the first network node to join the FL cluster (Paragraph [0140]: Since the other nodes in the first set of nodes 121 may be understood to also cast their respective votes of the leader node, the second node 112 may, in this Action 404, receive a respective vote from each of the other nodes in the first set of nodes 121 for, out of the nodes comprised in the first set of nodes 121, their respective candidate for leader node 113 of the first set of nodes 121. Paragraph [0142]: In this Action 405, the second node 112 may determine the third node 113 to be the leader node for the first set of nodes 121. The determining in this Action 405 may be based on the received respective votes in Action 404.)
Regarding claim 20, the combination of Banerjee and Jha teaches the apparatus of claim 19, wherein the processor is configured to (see rejection for claim 19);
Banerjee does not explicitly teach to provide the third message periodically or in response to an event trigger.
However, Jha teaches to provide the third message periodically or in response to an event trigger (Paragraph [0071]: Devices that satisfying requirements specified by the GA in the Periodic-DML-Participant-Solicitation-Request-Packet-P-1 may send a Periodic-DML-Participant-Response-Data-Packet-P-2 in response to the Periodic-DML-Participant-Solicitation-Request-Packet-P-1. Paragraph [0096]: The LC may maintain this data via updates pushed from the devices or via a periodical discovery broadcast or directed query of the devices. Also see paragraph [0095]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide the third message periodically or in response to an event trigger, as taught by Jha in the system of Banerjee, so that the local coordinator can send periodic information to the global coordinator (Jha: Paragraphs: [0071], [0095], [0096]).
Regarding claim 21, the combination of Banerjee and Jha teaches the apparatus of claim 19, wherein the processor is configured to (see rejection for claim 19);
Banerjee further teaches to provide the third message in a broadcast or in a groupcast to a plurality of network nodes including the first network node (Paragraph [0137]: In this Action 403, the second node 112 sends a second indication to the nodes in the first set of nodes 121. The second indication indicates the determined vote. Paragraph [0138]: The sending in this Action 403 may be implemented through a peer-to-peer, or broadcast, protocol.)
Regarding claim 22, the combination of Banerjee and Jha teaches the apparatus of claim 19 (see rejection for claim 19);
Banerjee further teaches wherein the third message includes second FL information of the apparatus (Paragraph [0084]: In order to eventually identify the clusters or sets of nodes, that may be useful to process data in order to predict the event, in this Action 301, the first node 111 may receive, respectively from each of the nodes in the first plurality of nodes 110, respective first information. According to a first option, the respective first information may be regarding: an arrival of service requests received over a first period of time by the first plurality of nodes 110 from a plurality of client computing devices 130 of the first plurality of nodes 110. A service request may be for example, a traffic condition service or a weather condition service. According to this option, the first node 111 may detect the pattern of arrival of services requests from clients. The first period of time may be, e.g., during training phase.)
Regarding claim 23, the combination of Banerjee and Jha teaches the apparatus of claim 19 (see rejection for claim 19);
Banerjee further teaches wherein the second message acknowledges an admission of the first network node to the FL cluster (Paragraph [0126]: In this Action 401 the second node 112 may receive the first indication from the first node 111 comprised in the communications system 100. Paragraph [0135]: In some embodiments, the determining in this Action 402 may be triggered by the receiving of the first indication in Action 401.)
Regarding claim 24, the combination of Banerjee and Jha teaches the apparatus of claim 12 wherein in response to the second message indicating the apparatus is the FL cluster leader of the FL cluster, the processor is further configured to: (see rejection for claim 12);
Banerjee does not explicitly teach to provide, to the first network node, a machine learning (ML) model configuration including an initial weight; obtain, from the first network node, a ML model information update including an update to the initial weight; and provide, to the first network node, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update of a third network node in the FL cluster.
However, Jha teaches to provide, to the first network node, a machine learning (ML) model configuration including an initial weight; obtain, from the first network node, a ML model information update including an update to the initial weight (Paragraph [0078]: For Example, the Participant-Solicitation-Request-Packet-B-1-LC may include one of more of the Tentative Round Information, DML Task Details, Communicate-Compute-Storage Requirements, query for Participant-Volume-Information, query for data quality, AoI, TEE requirement for training, or information to access the model training algorithm. Thus, the Participant-Solicitation-Request-Packet-B-1-LC may be thought of as a local version of Participant-Solicitation-Request-Packet-B-1, where locality is defined by the LC. Paragraph [0079]: Interested participant nodes-which may be limited to those that satisfy the requirements specified by the GA-respond with a Participant-Response-Data-Packet-B-2 data packet to the LC (message(s) 715). The Participant-Response-Data-Packet-B-2 identifies the node's intent to participate in the DML training (e.g., a current DML training round). The Participant-Response-Data-Packet-B-2 may include one or more of Availability-Windows for the node, Communicate-Compute-Storage Capabilities of the node; or Data-quality of the node's data that will be used to training the model. In an example, if the device intends to participate in the DML round and does not have the model training algorithm, the device may retrieve the model training algorithm prior to sending the Participant-Response-Data-Packet-B-2.)
and provide, to the first network node, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update of a third network node in the FL cluster (Paragraph [0085]: In response to receiving the Model-Download-Data-Packet-B-3, the LC may send a Model-Download-Data-Packet-B-3-LC (message 730) data packet to the selected participant devices. The LC may transmit the Model-Download-Data-Packet-B-3-LC as broadcast (e.g., if several participants are selected) or unicast (e.g., if only one or two participants selected from this LC) message. Paragraph [0086]: At this stage, the participant devices perform model training in accordance with the DML training round criteria (e.g., using the specified model training algorithm over a specified set of data, etc.) to produce an update for the model. The devices may then transmit the produce model update to the LC in a Model-Update-Data-Packet-B-4 data packet (message 735). In an example, the Model-Update-Data-Packet-B-4 data packet is a unicast message to the LC.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide, to the first network node, a machine learning (ML) model configuration including an initial weight; obtain, from the first network node, a ML model information update including an update to the initial weight; and provide, to the first network node, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update of a third network node in the FL cluster, as taught by Jha in the system of Banerjee, so that model updates can be performed by the participant devices, and forwarded to the global aggregator by the local coordinator (Jha: Paragraphs [0078], [0079], [0085], [0086]).
Regarding claim 25, the combination of Banerjee and Jha teaches the apparatus of claim 24 (see rejection for claim 24);
Banerjee does not explicitly teach wherein the aggregated ML model information update further includes a second aggregation of the ML model information update with a third ML model information update of a fourth network node in a second FL cluster.
However, Jha teaches wherein the aggregated ML model information update further includes a second aggregation of the ML model information update with a third ML model information update of a fourth network node in a second FL cluster (Paragraph [0084]: After participant devices are selected, the GA may send a Model-Download-Data-Packet-B-3 (message 725) data packet for this round of DML to each LC. In an example, the Model-Download-Data-Packet-B-3 is unicast to the LCs. In an example, the Model-Download-Data-Packet-B-3 includes a selected participant list if, for example, the participants are selected by the GA or if the GA has overridden one or more of an LC's participant device selections. Paragraph [0086]: At this stage, the participant devices perform model training in accordance with the DML training round criteria (e.g., using the specified model training algorithm over a specified set of data, etc.) to produce an update for the model. The devices may then transmit the produce model update to the LC in a Model-Update-Data-Packet-B-4 data packet (message 735). In an example, the Model-Update-Data-Packet-B-4 data packet is a unicast message to the LC. Paragraph [0087]: Once the aggregated model update is produced, the LC transmits a Model-Update-Data-Packet-B-4-LC (message 740) data packet to the GA. This Model-Update-Data-Packet-B-4-LC includes the aggregated model update. In an example, the Model-Update-Data-Packet-B-4-LC is a unicast message.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide wherein the aggregated ML model information update further includes a second aggregation of the ML model information update with a third ML model information update of a fourth network node in a second FL cluster, as taught by Jha in the system of Banerjee, so that model updates can be performed by the participant devices, and forwarded to the global aggregator by the local coordinator (Jha: Paragraphs [0084], [0086], [0087]).
Regarding claim 26, the combination of Banerjee and Jha teaches the apparatus of claim 12 (see rejection for claim 12);
Banerjee further teaches wherein the second message indicates the second network node is the FL cluster leader (Paragraph [0145]: In this Action 406, the second node 112 may exchange a third indication with the nodes in the first set of nodes 121. The third indication may indicate that the third node 113 is the leader node 113 for the first set of nodes 121. The exchanging of the third indication may be based on the sent second indication in Action 403.)
Regarding claim 27, the combination of Banerjee and Jha teaches the apparatus of claim 26 wherein the processor is configured to (see rejection for claim 26);
Banerjee does not explicitly teach to obtain the first message from the second network node; and wherein the first message further includes an identifier of the first network node.
However, Jha teaches to obtain the first message from the second network node; and wherein the first message further includes an identifier of the first network node (Paragraph [0080]: In an example, the response packets may also include device attestation evidence according to trustworthiness properties of the device. A device may employ a variety of attestation roots of trust including but not limited to a trusted platform module (TPM), a device identifier composition engine (DICE), software guard extensions (SGX), trusted domain extension (TDX), TrustZone, etc. In an example, the LC or the GC may act as attestation verifiers for provided attestation evidence. In an example, the LC or the GC may have security policies that require verification of attestation evidence as a pre-condition to acting on or forwarding other information contained in the message. In an example, the DML round context (e.g., the Round Start time or the Round End time) may include a session handle or context identifier.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing
date of the claimed invention to obtain the first message from the second network node; and wherein the first message further includes an identifier of the first network node, as taught by Jha in the system of Banerjee, since security policies require verification of the devices before forwarding other information contained in the message (Jha: Paragraph [0080]).
Regarding claim 28, the combination of Banerjee and Jha teaches the apparatus of claim 12, wherein the processor is further configured to: (see rejection for claim 12);
Banerjee does not explicitly teach to obtain, from the second network node, a machine learning (ML) model information update; and provide, to the second network node, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update from a third network node in a second FL cluster.
However, Jha teaches to obtain, from the second network node, a machine learning (ML) model information update (Paragraph [0078]: For Example, the Participant-Solicitation-Request-Packet-B-1-LC may include one of more of the Tentative Round Information, DML Task Details, Communicate-Compute-Storage Requirements, query for Participant-Volume-Information, query for data quality, AoI, TEE requirement for training, or information to access the model training algorithm. Thus, the Participant-Solicitation-Request-Packet-B-1-LC may be thought of as a local version of Participant-Solicitation-Request-Packet-B-1, where locality is defined by the LC. Paragraph [0079]: Interested participant nodes-which may be limited to those that satisfy the requirements specified by the GA-respond with a Participant-Response-Data-Packet-B-2 data packet to the LC (message(s) 715). The Participant-Response-Data-Packet-B-2 identifies the node's intent to participate in the DML training (e.g., a current DML training round). The Participant-Response-Data-Packet-B-2 may include one or more of Availability-Windows for the node, Communicate-Compute-Storage Capabilities of the node; or Data-quality of the node's data that will be used to training the model. In an example, if the device intends to participate in the DML round and does not have the model training algorithm, the device may retrieve the model training algorithm prior to sending the Participant-Response-Data-Packet-B-2.)
and provide, to the second network node, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update from a third network node in a second FL cluster (Paragraph [0085]: In response to receiving the Model-Download-Data-Packet-B-3, the LC may send a Model-Download-Data-Packet-B-3-LC (message 730) data packet to the selected participant devices. The LC may transmit the Model-Download-Data-Packet-B-3-LC as broadcast (e.g., if several participants are selected) or unicast (e.g., if only one or two participants selected from this LC) message. Paragraph [0086]: At this stage, the participant devices perform model training in accordance with the DML training round criteria (e.g., using the specified model training algorithm over a specified set of data, etc.) to produce an update for the model. The devices may then transmit the produce model update to the LC in a Model-Update-Data-Packet-B-4 data packet (message 735). In an example, the Model-Update-Data-Packet-B-4 data packet is a unicast message to the LC.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to obtain, from the second network node, a machine learning (ML) model information update; and provide, to the second network node, an aggregated ML model information update including an aggregation of the ML model information update with a second ML model information update from a third network node in a second FL cluster, as taught by Jha in the system of Banerjee, so that model updates can be performed by the participant devices, and forwarded to the global aggregator by the local coordinator (Jha: Paragraphs [0078], [0079], [0085], [0086]).
Regarding claim 29, Banerjee teaches a method of wireless communication at a user equipment (UE), comprising: providing a first message including federated Learning (FL) information; providing a second message indicating a first network node; and obtaining a third message indicating one of the first network node or a second network node as an FL cluster leader and indicating an FL cluster of the UE based on the FL information (see rejection for claim 1);
Banerjee does not explicitly teach a method at a user equipment (UE); indicating an FL cluster of
the UE based on the FL information.
However, Jha teaches a method at a user equipment (UE); indicating an FL cluster of the UE based on the FL information (Paragraph [0046]: Generally, the examples assume that the data used for training is stored at the participant devices, such as mobile phones, tablets, vehicles, home sensors. Also see rejection for claim 1);
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide indicating an FL cluster of the UE based on the FL information, as taught by Jha in the system of Banerjee, so that the local coordinator can indicate to the participating nodes that they are part of the cluster (Jha: Paragraph [0077]).
Regarding claim 30, Banerjee teaches a method of wireless communication at a first network node, comprising: obtaining a first message including federated learning (FL) information of a second network node; and providing a second message indicating an FL cluster of the second network node based on the FL information, wherein the second message indicates one of the first network node or a third network node as an FL cluster leader of the FL cluster (see rejection for claim 1);
Banerjee does not explicitly teach indicating an FL cluster of the second network node based on
the FL information.
However, Jha teaches indicating an FL cluster of the second network node based on
the FL information (see rejection for claim 1);
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing
date of the claimed invention to provide indicating an FL cluster of the second network node based on
the FL information, as taught by Jha in the system of Banerjee, so that the local coordinator can indicate to the participating nodes that they are part of the cluster (Jha: Paragraph [0077]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LATHA CHAKRAVARTHY whose telephone number is (703)756-1172. The examiner can normally be reached M-Th 8:30 AM - 5 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, Huy Vu can be reached at 571-272-3155. 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.
/L.C./Examiner, Art Unit 2461
/JASON E MATTIS/Primary Examiner, Art Unit 2461
