CN113159190A - Federal incentive distribution method and device, computer equipment and storage medium - Google Patents
Federal incentive distribution method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113159190A CN113159190A CN202110449555.4A CN202110449555A CN113159190A CN 113159190 A CN113159190 A CN 113159190A CN 202110449555 A CN202110449555 A CN 202110449555A CN 113159190 A CN113159190 A CN 113159190A
- Authority
- CN
- China
- Prior art keywords
- federal
- participant
- determining
- incentive
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 130
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 213
- 239000013598 vector Substances 0.000 claims abstract description 76
- 230000005284 excitation Effects 0.000 claims abstract description 63
- 238000000691 measurement method Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims description 30
- 238000004891 communication Methods 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 14
- 230000005540 biological transmission Effects 0.000 claims description 13
- 238000004140 cleaning Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000005304 joining Methods 0.000 claims description 4
- 230000008901 benefit Effects 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 6
- 239000000047 product Substances 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for allocating joint incentives, computer equipment and a storage medium, wherein effective training data sets corresponding to all participants and training quality vectors thereof are determined by sending initial training data sets by all participants who self-join a joint system; determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to all participants and the total amount of effective data; determining a contribution value of each participant by adopting a marginal utility measurement method, and determining a preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant; determining actual incentive distribution values corresponding to all the parties according to the preset incentive distribution values and a preset incentive determining strategy; and executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant. The invention improves the comprehensive benefit of the federal system.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a device for allocating federated incentives, computer equipment and a storage medium.
Background
The federated learning technology has the advantages of distributed machine learning and privacy protection technology, and can be used for training in a combined manner in multiple ways on the premise of ensuring data safety and privacy, so that the model performance and the actual benefit are improved.
In the prior art, the application scenario of federal learning is premised on active participation of multiple participants and training of local models using high quality data. However, since the quality and quantity of the federally learned input data are determined by the participants, the problem may occur that the allocation of federal learning incentives in the federal system cannot be matched with the requirements of each participant, which in turn results in lower overall efficiency of the federal system.
Disclosure of Invention
The embodiment of the invention provides a federated incentive allocation method, a federated incentive allocation device, computer equipment and a storage medium, and aims to solve the problem of low comprehensive benefit of a federated system.
A method of federated incentive assignment, comprising:
receiving an initial training data set sent by each participant joining a federated system, and determining an effective training data set corresponding to each participant and a training quality vector thereof from the initial training data set; associating a total amount of valid data with a valid training data set of said participant;
determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to the participants and the total effective data amount;
determining a contribution value of each participant by adopting a marginal utility measurement method, and determining a preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant;
determining actual incentive distribution values corresponding to all the parties according to the preset incentive distribution values and a preset incentive determining strategy;
and executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
A federated incentive dispensing device, comprising:
the system comprises a data processing module, a data processing module and a data processing module, wherein the data processing module is used for receiving an initial training data set sent by each participant who joins a federated system, and determining an effective training data set corresponding to each participant and a training quality vector thereof from the initial training data set; associating a total amount of valid data with a valid training data set of said participant;
the federal depth of excitation determination module is used for determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to the participants and the total amount of effective data;
the preset incentive distribution value determining module is used for determining the contribution value of each participant by adopting a marginal utility measurement method, and determining the preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant;
the actual incentive distribution value determining module is used for determining an actual incentive distribution value corresponding to each party according to the preset incentive distribution value and a preset incentive determining strategy;
and the incentive distribution task execution module is used for executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the federal incentive distribution method when executing the computer program.
A computer-readable storage medium storing a computer program which, when executed by a processor, implements the federal incentive distribution method described above.
According to the federal incentive distribution method, the apparatus, the computer device and the storage medium, an effective training data set and a training quality vector thereof corresponding to each participant are determined from an initial training data set by receiving the initial training data set sent by each participant added into a federal system; associating a total amount of valid data with a valid training data set of said participant; determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to the participants and the total effective data amount; determining a contribution value of each participant by adopting a marginal utility measurement method, and determining a preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant; determining actual incentive distribution values corresponding to all the parties according to the preset incentive distribution values and a preset incentive determining strategy; and executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
According to the method, the contribution of each participant to the federal training of the federal system is evaluated through the effective data total amount of the effective training data set transmitted by each participant and the training quality vector corresponding to the effective training data set, so that the incentive matched with the contribution value corresponding to each participant is determined, and the preset incentive determining strategy is introduced, so that all participants have positive incentive, more participants can be attracted to provide more training data with better quality, the training data are added into the federal system, and the comprehensive benefit of the federal system is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a schematic diagram of an application environment of the federal incentive distribution method in an embodiment of the present invention;
FIG. 2 is a flow diagram of a federated incentive distribution method in one embodiment of the present invention;
FIG. 3 is a flowchart of step S10 of the federal incentive distribution method in an embodiment of the present invention;
FIG. 4 is a flowchart of step S20 of the federal incentive distribution method in an embodiment of the present invention;
FIG. 5 is a flowchart of step S40 of the federal incentive distribution method in an embodiment of the present invention;
FIG. 6 is a functional block diagram of a federated incentive distribution facility in one embodiment of the present invention;
FIG. 7 is a functional block diagram of the data processing module of the Federal incentive distribution device in an embodiment of the present invention;
FIG. 8 is a functional block diagram of a federated incentive depth determination module in a federated incentive distribution device in an embodiment of the present invention;
FIG. 9 is a schematic block diagram of an actual incentive distribution value determining module in the federal incentive distribution device in an embodiment of the present invention;
FIG. 10 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The federal incentive distribution method provided by the embodiment of the invention can be applied to the application environment shown in fig. 1. Specifically, the federal incentive distribution method is applied to a federal incentive distribution system, which includes a client and a server as shown in fig. 1, wherein the client and the server communicate through a network, so as to solve the problem of low comprehensive benefit of the federal system. The client is also called a user side, and refers to a program corresponding to the server and providing local services for the client. The client may be installed on, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server may be implemented as a stand-alone server or as a server cluster consisting of a plurality of servers.
In one embodiment, as shown in fig. 2, a method for allocating federated incentives is provided, which is described by taking the server in fig. 1 as an example, and includes the following steps:
s10: receiving an initial training data set sent by each participant joining a federated system, and determining an effective training data set corresponding to each participant and a training quality vector thereof from the initial training data set; an active training data set of the participant is associated with an amount of active data.
As will be appreciated, participant refers to a user or terminal determined to participate in the federal system training. A federal system refers to a system based on federal learning techniques and awaiting federal training. The initial training data set refers to local data of each participant, that is, data that each participant determines to input to use for federal system training. The valid training data set refers to a set of the initial training data set excluding the remaining training data that do not meet the training requirements, and the training data in the valid training data set all meet the training requirements of the federated system. The training quality vector is used for representing the quality of training data in the effective training data set, and is obtained through various dimensional performance evaluations such as saturation, similarity and the like. The total amount of valid data refers to the total number of valid training data in each valid training data set.
In one embodiment, as shown in fig. 3, the initial training data set includes at least one initial training data; step S10 includes:
s101: receiving a data cleaning instruction containing a training requirement, and performing data cleaning processing on the initial training set of each participant to remove initial training data which does not meet the training requirement in the initial training set.
S102: and recording the initial training set after the initial training data which do not meet the training requirement are removed as the effective training data set.
As will be appreciated, training requirements refer to the requirement for federal training of federal systems, which may include requirements for training data, requirements for model parameters, and the like. The data cleansing instructions may be sent by a user (e.g., federal system trainers) or may be automatically generated after entering training requirements.
Specifically, after a data cleaning instruction containing a training requirement is received, data cleaning processing is performed on initial training data in initial training sets of all participants to remove initial training data which do not meet the training requirement in the initial training data sets, and the initial training sets after the initial training data which do not meet the training requirement are removed are recorded as effective training data sets. Further, the total amount of valid data associated with the valid training data set is a difference between the total amount of initial training data in the initial training data set and the total amount of initial training data that does not meet the training requirement.
S103: inputting the effective training data sets into a federal feature engineering module, performing feature evaluation processing on the effective training data sets through the federal feature engineering module, and determining training quality vectors corresponding to the effective training data sets.
As may be appreciated, a federated feature engineering module refers to a module in a federated system that measures the quality of data in an effective training dataset from a number of different dimensions.
Specifically, after an initial training set from which initial training data not meeting the training requirements are removed is recorded as the effective training data set, the effective training data set is input into a federal feature engineering module in a federal system, and the characteristic evaluation processing is performed on the training data in the effective training data set through the federal feature engineering module from multiple dimensions such as saturation, sparsity, similarity, data distribution and the like, so as to determine training quality vectors corresponding to the effective training data sets.
S20: and determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to all the participants and the total amount of effective data.
It can be understood that after each participant provides data for federal training of the federal system, the participant is subjected to incentive feedback, and the federal incentive depth is a parameter affecting incentive feedback on the participant, and the larger the federal incentive depth is, the more incentive feedback is given to each participant; on the contrary, the smaller the federal excitation depth is, the less excitation feedback is given to each participant.
In one embodiment, as shown in fig. 4, step S20 includes:
s201: determining a matching quality vector from the training quality vectors corresponding to each of the participants; the matching quality refers to a training quality vector with the highest matching degree with the training requirement.
It can be understood that after the feature evaluation processing is performed on the effective training data sets through the federal feature engineering module to determine the training quality vectors corresponding to the effective training data sets, the training quality vector which is most matched with the training requirement, that is, the training quality vector with the highest quality can be determined according to the training quality vectors corresponding to the effective training data sets corresponding to the participants, and then the training quality vector with the highest quality is recorded as the matching quality vector.
S202: and determining an average quality vector by adopting a mathematical expectation algorithm according to the training quality vectors corresponding to all the participants.
Specifically, after the effective training data sets are subjected to feature evaluation processing through a federal feature engineering module, and training quality vectors corresponding to the effective training data sets are determined, average quality vectors are determined through a mathematical expectation algorithm according to the training quality vectors corresponding to the participants. Illustratively, the average quality vector may be determined by the following mathematical expectation algorithm:
wherein the content of the first and second substances,
mean quality vectors corresponding to m participants; e (qi) refers to the training quality vector corresponding to the ith participant; m is the total number of participants.
S203: and determining an average effective data amount by adopting a mathematical expectation algorithm according to the total effective data amount corresponding to each participant.
Specifically, after the initial training set from which the initial training data not meeting the training requirement is removed is recorded as the effective training data set, the number of training data in each effective training data set, that is, the total effective data amount of the effective training data set, is determined, and then the average effective data amount is determined by using a mathematical expectation algorithm according to the total effective data amount associated with the effective training data sets corresponding to the participants. Illustratively, the average effective data amount may be determined by the following mathematical expectation algorithm:
wherein the content of the first and second substances,
means the average effective data amount corresponding to m participants; q (i) refers to the total amount of valid data for the ith participant.
S204: and acquiring the maximum data carrying capacity of the federal system, and determining the federal excitation depth according to the matching quality vector, the average effective data and the maximum data carrying capacity.
It is understood that maximum data carrying capacity refers to the maximum amount of training data that the federated system can carry. And then after the maximum data carrying capacity of the federal system is obtained, determining the federal depth of excitation according to the matching quality vector, the average effective data and the maximum data carrying capacity.
In one embodiment, step S204 includes:
receiving system service parameters sent by each participant, and determining total service parameters of the federated system according to the system service parameters of each participant.
It is understood that the system service parameter refers to the system service fee that each participant needs to submit after determining to join the federal system for federal training (the system service fee submitted by each participant can be set to the same fee). The total service parameter is the sum of the system service fees submitted by all the participants.
Specifically, after receiving the system service parameters sent by each participant, the sum of the system service parameters of each participant is recorded as the total service parameter of the federal system.
Acquiring a first preset quantity decision parameter, a second preset quantity decision parameter, a first preset depth decision parameter and a second preset depth decision parameter of the federal system; the second predetermined number decision parameter is greater than the first predetermined number decision parameter.
It can be understood that the first preset number decision parameter, the second preset number decision parameter, the first preset depth decision parameter and the second preset depth decision parameter are decision parameters of a federal system, and the parameters can be determined by various factors such as an application environment of the federal system and a system operation condition. The first predetermined number decision parameter and the second predetermined number decision parameter are used to measure the average effective data amount. The first preset depth decision parameter and the second preset depth decision parameter are used for determining the size of the federal depth of excitation.
And when the average effective data volume is smaller than the first preset number decision parameter, determining the federal depth of excitation according to the total service parameters, the matching quality vectors, the average effective data and the maximum data carrying capacity.
Specifically, after a first preset quantity decision parameter, a second preset quantity decision parameter, a first preset depth decision parameter and a second preset depth decision parameter of the federal system are obtained, the average effective data volume is compared with the first preset quantity decision parameter and the second preset quantity decision parameter, and when the average effective data volume is smaller than the first preset quantity decision parameter, the federal excitation depth is determined according to the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity. For example, the corresponding federal depth of excitation when the average effective data volume is less than the first preset number decision parameter may be determined according to the following expression:
wherein, T1 is the corresponding federal depth of excitation when the average effective data volume is less than the first predetermined number decision parameter; c is a system service parameter corresponding to the participant (here, the system service parameter corresponding to each participant is set as C, and if the system service parameters corresponding to the participants are different, the sum of the system service parameters corresponding to each participant can be replaced by C);
mean quality vectors corresponding to m participants;
means the average effective data amount corresponding to m participants;
is a matching quality vector;
is the maximum data carrying capacity; x1 is a first predetermined number decision parameter.
And when the average effective data volume is greater than or equal to the first preset number decision parameter and less than the second preset number decision parameter, determining the federal excitation depth according to the first preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity.
Specifically, after a first preset quantity decision parameter, a second preset quantity decision parameter, a first preset depth decision parameter and a second preset depth decision parameter of the federal system are obtained, the average effective data volume is compared with the first preset quantity decision parameter and the second preset quantity decision parameter, and when the average effective data volume is greater than or equal to the first preset quantity decision parameter and smaller than the second preset quantity decision parameter, the federal excitation depth is determined according to the first preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity. For example, the corresponding federal depth of excitation when the average effective data volume is greater than or equal to the first preset number decision parameter and less than the second preset number decision parameter may be determined according to the following expression:
wherein, T2 is the corresponding federal depth of excitation when the average effective data volume is greater than or equal to the first predetermined number decision parameter and less than the second predetermined number decision parameter; x2 is a second predetermined number decision parameter; t1 is a first preset depth decision parameter.
And when the average effective data volume is greater than or equal to the second preset number decision parameter, determining the federal depth of excitation according to the second preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity.
Specifically, after a first preset number decision parameter, a second preset number decision parameter, a first preset depth decision parameter and a second preset depth decision parameter of the federal system are obtained, the average effective data volume is compared with the first preset number decision parameter and the second preset number decision parameter, and when the average effective data volume is greater than or equal to the second preset number decision parameter, the federal excitation depth is determined according to the second preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity. For example, the corresponding federal depth of excitation when the average amount of useful data is greater than or equal to the second predetermined number decision parameter may be determined according to the following expression:
wherein, T3 is the corresponding federal depth of excitation when the average effective data volume is greater than or equal to the second predetermined number decision parameter; t2 is a second preset depth decision parameter.
S30: determining the contribution value of each participant by adopting a marginal utility measurement method, and determining a preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant.
It will be appreciated that the marginal utility measure is used to measure the contribution each participant makes to the training of the federated system. The preset incentive assignment value refers to a value for assigning incentives to each participant in advance according to the participant's contribution to the training of the federated system.
In one embodiment, step S30 includes:
and determining the marginal utility of each participant on the federal system by adopting a Shapley value algorithm according to the effective training data set corresponding to each participant.
And determining a contribution value corresponding to each participant according to the marginal utility corresponding to each participant.
Among other things, the sharley value algorithm is used to measure the contribution of each participant to the training of the federated system. Specifically, after the federal depth of incentive of the federal system is determined according to training quality vectors and the total amount of effective data corresponding to each participant, a Shapley value algorithm is adopted to determine marginal utility of each participant on the federal system in the training process of the federal system, wherein the marginal utility is brought by the effective training data set corresponding to each participant, and contribution values corresponding to each participant are determined according to the marginal utility corresponding to each participant. Further, the sum of the contribution values corresponding to each participant is 1.
Further, the contribution value corresponding to each of the parties may be determined by the following expression:
δi＝v(S∪{i})-v(S)
wherein, δ i is the marginal utility brought by the addition of the ith participant to the federal system; v (S U { i }) is marginal utility brought by all participants after being added into the federal system; v (S) marginal utility brought by other participants except the ith participant after joining the federal system (S is a set not including the ith participant);
the contribution value corresponding to each participant; m is the set of all participants; m is the number of participants.
Exemplarily, assuming that there are 2 participants X and Y in total, if there is only a participant X, the marginal utility corresponding to the federal system at this time is v (X); if the participation party X and the participation party Y exist and the marginal utility corresponding to the federal system is v (X + Y), the marginal utility corresponding to the participation party Y is v (X + Y) -v (X); assuming that there are a total of 3 parties X, Y, Z, if the marginal utility of party X needs to be calculated, the set of all parties is enumerated as X, Y, Z, X, Y, X, Z, Y, Z, where the subset of parties a is excluded as Y, Z, then these subsets can be represented by S.
S40: and determining an actual incentive distribution value corresponding to each party according to the preset incentive distribution value and a preset incentive determining strategy.
It will be appreciated that the predictive incentive determination strategy is used to determine the actual incentive distribution values corresponding to each participant. Since the participants generate data calculation and communication loss in the federal training process, when the incentive distributed to the participants is less, the total consumption caused by the data calculation and the communication loss may not be met by the incentive actually obtained by the participants, so that the predicted incentive determining strategy is introduced in the embodiment, so that when the preset incentive distribution value of the participant does not meet the total consumption caused by the data calculation and the communication loss, the corresponding preset incentive distribution value is supplemented to the incentive distribution value matched with the total consumption through an incentive pool in the federal system.
In one embodiment, as shown in fig. 5, step S40 includes:
s401: obtaining a base loss value corresponding to each of the participants and comparing the default incentive distribution value corresponding to the same participant with the base loss value.
As will be appreciated, the base loss value refers to the sum of the data calculations and communication losses that the participants made during the federal training process.
In an embodiment, before step S401, the method includes:
and determining a calculated loss value corresponding to each participant according to the hardware equipment parameters corresponding to each participant and the total effective data amount through a calculated loss function.
It is understood that the hardware device parameters may be capacitance coefficient of the terminals of the participating parties, Processing cycle number of a CPU (Central Processing Unit) of the terminals of the participating parties, Processing cycle frequency, and the like.
Specifically, the calculated loss value corresponding to each of the parties may be determined by the following expression:
wherein Ei is a calculated loss value corresponding to the ith participant;
the capacitance coefficient of the terminal that is the party; ci is the total amount of effective data corresponding to the ith participant; di is the number of Processing cycles of the CPU (Central Processing Unit) of the participant's terminal; fi is a Processing cycle frequency of a CPU (Central Processing Unit) of the terminal of the participant.
Determining, by a communication loss function, a communication loss value corresponding to each of the participants based on the communication transmission parameter corresponding to each of the participants.
It is understood that the communication transmission parameters may include data transmission duration, transmission power, total amount of transmission data, network bandwidth, etc.
Specifically, the communication loss value corresponding to each of the parties may be determined by the following expression:
Fi＝τipi(si/τi)
fi is a communication loss value corresponding to the ith participant; τ i is the transmission duration for the ith participant to transmit the initial training data set; pi is the transmission power of the ith participant to transmit the initial training data set; si is the total amount of initial training data in the initial training data set transmitted by the ith participant; n0 is transmission background noise; hi is the terminal channel gain of the ith participant; and B is the network bandwidth.
Determining, by a product logarithm function, a loss cost corresponding to each of the participants based on the hardware device parameters corresponding to each of the participants and the communication transmission parameters.
Wherein the logarithmic function of product is a Lambert W function, and the logarithmic function of product is used to determine the loss cost corresponding to each participant. It can be understood that, due to the hardware heterogeneity of the terminal of each participant, the loss generated by each participant is different, and therefore, by introducing the loss cost of each participant, the accuracy of the determined base loss value corresponding to each participant can be improved.
Specifically, the loss cost corresponding to each of the participants may be determined by the following expression:
wherein gi is a loss cost corresponding to the ith participant; w () is a product logarithm function; k is a loss parameter that can be changed according to the state of the CPU of the terminal of each participant.
Determining a base loss value corresponding to each of the participants based on the calculated loss value, the communication loss value, and the loss cost corresponding to each of the participants.
Specifically, after the calculated loss value, the communication loss value, and the loss cost corresponding to each participant are determined, the base loss value corresponding to each participant is determined according to the calculated loss value, the communication loss value, and the loss cost corresponding to each participant. Further, the base loss value corresponding to each participant may be determined according to the following expression:
ri is a basic loss value corresponding to the ith participant;
for the optimal service capability value corresponding to the ith participant, the optimal service capability value may be determined according to the loss parameters k in the above description (understandably, for different terminals, the optimal service capability value corresponding to each loss parameter k has been preset, and the mapping relationship is stored in the preset storage table, so that after determining the loss parameter k, the optimal service capability value may be determined by querying the preset storage table).
S402: and when the preset excitation distribution value is larger than or equal to the basic loss value, recording the actual excitation distribution value of the participant corresponding to the preset excitation distribution value as the preset excitation distribution value.
Specifically, after obtaining the basic loss value corresponding to each participant, comparing the preset incentive distribution value corresponding to the same participant with the basic loss value, and when the preset incentive distribution value is greater than or equal to the basic loss value, indicating that the total amount of the effective training numbers transmitted by the participant is large and the quality of the training data is high, thus recording the preset incentive distribution value as the actual incentive distribution value corresponding to the participant.
S403: and when the preset excitation distribution value is smaller than the basic loss value, recording the actual excitation distribution value of the participant corresponding to the basic loss value as the basic loss value according to the preset excitation determination strategy.
Specifically, after obtaining the basic loss values corresponding to the participants, comparing the preset incentive distribution value corresponding to the same participant with the basic loss value, and when the preset incentive distribution value is smaller than the basic loss value, indicating that the total amount of the effective training numbers transmitted by the participant is smaller and the quality of the training data is lower, so that the preset incentive distribution value is supplemented through an incentive pool of the federal system through a preset incentive determining strategy to supplement the preset incentive distribution value to the basic loss value, and recording the basic loss value as the actual incentive distribution value corresponding to the participant.
S50: and executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
Specifically, after the actual incentive distribution value corresponding to each participant is determined according to the preset incentive distribution value and the preset incentive determining strategy, the federal incentive distribution task is executed according to the actual incentive distribution value corresponding to each participant, so that the actual incentive distribution value corresponding to each participant is distributed to each participant.
In this embodiment, the contribution of each participant to the federal training of the federal system is evaluated by the total amount of effective data of the effective training data set transmitted by each participant and the training quality vector corresponding to the effective training data set to determine the incentive matched with the contribution value corresponding to each participant, and a preset incentive determining strategy is introduced, so that the participant with less contribution to the federal training of the federal system can also obtain the incentive matched with the basic consumption value thereof, and further all participants have forward incentives, so that more participants can be attracted to provide more training data with better quality, and the comprehensive benefit of the federal system is improved.
In another embodiment, to ensure the privacy and security of the initial training data set in the above embodiments, the initial training data set may be stored in a blockchain. The Block chain (Blockchain) is an encrypted and chained transaction storage structure formed by blocks (blocks).
For example, the header of each block may include hash values of all transactions in the block, and also include hash values of all transactions in the previous block, so as to achieve tamper resistance and forgery resistance of the transactions in the block based on the hash values; newly generated transactions, after being filled into the tiles and passing through the consensus of nodes in the blockchain network, are appended to the end of the blockchain to form a chain growth.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In an embodiment, a federal incentive distribution device is provided, and the federal incentive distribution device corresponds to the federal incentive distribution method in the above embodiments one to one. As shown in fig. 6, the federal incentive distribution device includes a data processing module 10, a federal incentive depth determination module 20, a preset incentive distribution value determination module 30, an actual incentive distribution value determination module 40, and an incentive distribution task execution module 50. The functional modules are explained in detail as follows:
the system comprises a data processing module 10, a quality vector determining module and a quality vector determining module, wherein the data processing module is used for receiving an initial training data set sent by each participant who joins in a federal system, and determining an effective training data set and a training quality vector thereof corresponding to each participant from the initial training data set; associating a total amount of valid data with a valid training data set of said participant;
a federal depth of excitation determination module 20, configured to determine a federal depth of excitation of the federal system according to the training quality vectors and the total amount of effective data corresponding to each of the participants;
a preset incentive distribution value determining module 30, configured to determine a contribution value of each participant by using a marginal utility measurement method, and determine a preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant;
an actual incentive distribution value determining module 40, configured to determine, according to the preset incentive distribution value and a preset incentive determining policy, an actual incentive distribution value corresponding to each of the parties;
and the incentive distribution task execution module 50 is used for executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
Preferably, as shown in fig. 7, the initial training data set includes at least one initial training data; the data processing module 10 comprises:
a data cleaning unit 101, configured to receive a data cleaning instruction including a training requirement, to perform data cleaning processing on the initial training set of each participant, so as to remove initial training data in the initial training set that does not meet the training requirement;
an effective data determining unit 102, configured to record, as the effective training data set, an initial training set from which the initial training data that does not meet the training requirement is removed;
and the feature evaluation unit 103 is configured to input the effective training data sets into a federal feature engineering module, perform feature evaluation processing on the effective training data sets through the federal feature engineering module, and determine training quality vectors corresponding to the effective training data sets.
Preferably, as shown in fig. 8, the federal depth of excitation determination module 20 includes:
a matching quality vector determination unit 201 for determining a matching quality vector from the training quality vectors corresponding to the respective participants; the matching quality refers to a training quality vector with the highest matching degree with the training requirement;
an average quality vector determining unit 202, configured to determine an average quality vector by using a mathematical expectation algorithm according to the training quality vectors corresponding to the participants;
an average effective data amount determining unit 203, configured to determine an average effective data amount by using a mathematical expectation algorithm according to the total effective data amount corresponding to each of the participants;
and the federal depth of excitation determining unit 204 is configured to obtain a maximum data carrying capacity of the federal system, and determine the federal depth of excitation according to the matching quality vector, the average effective data, and the maximum data carrying capacity.
Preferably, the federal depth of excitation determination unit 204 includes:
a total service parameter determining subunit, configured to receive the system service parameters sent by each of the participants, and determine a total service parameter of the federated system according to the system service parameters of each of the participants;
the parameter obtaining subunit is configured to obtain a first preset quantity decision parameter, a second preset quantity decision parameter, a first preset depth decision parameter, and a second preset depth decision parameter of the federal system; the second predetermined number decision parameter is greater than the first predetermined number decision parameter;
a first federal depth of excitation determination subunit, configured to determine the federal depth of excitation according to the total service parameters, the matching quality vectors, the average effective data, and the maximum data carrying capacity when the average effective data amount is smaller than the first preset number decision parameter;
a second federated excitation depth determination subunit, configured to determine the federated excitation depth according to the first preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data, and the maximum data carrying capacity when the average effective data amount is greater than or equal to the first preset number decision parameter and smaller than the second preset number decision parameter;
and the third federal excitation depth determining subunit is used for determining the federal excitation depth according to the second preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity when the average effective data volume is greater than or equal to the second preset number decision parameter.
Preferably, the preset excitation assignment value determination module 30 includes:
the marginal utility determining unit is used for determining the marginal utility of each participant on the federal system by adopting a Shapley value algorithm according to the effective training data set corresponding to each participant;
and the contribution value determining unit is used for determining the contribution value corresponding to each participant according to the marginal utility corresponding to each participant.
Preferably, as shown in fig. 9, the actual stimulus allocation value determination module 40 includes:
a fundamental loss value obtaining unit 401, configured to obtain a fundamental loss value corresponding to each of the participants, and compare the preset excitation allocation value corresponding to the same participant with the fundamental loss value;
a first actual incentive distribution value determining unit 402, configured to record, as the preset incentive distribution value, an actual incentive distribution value of the participant corresponding to the preset incentive distribution value when the preset incentive distribution value is greater than or equal to the base loss value;
a second actual excitation allocation value determining unit 403, configured to record, as the basic loss value, the actual excitation allocation value of the participant corresponding to the basic loss value according to the preset excitation determination policy when the preset excitation allocation value is smaller than the basic loss value.
Preferably, the federal incentive distribution device further comprises:
a calculation loss value determining module, configured to determine, by calculating a loss function, a calculation loss value corresponding to each of the participants according to the hardware device parameters corresponding to each of the participants and the total amount of the valid data;
a communication loss value determining module, configured to determine, according to a communication transmission parameter corresponding to each of the participants, a communication loss value corresponding to each of the participants through a communication loss function;
a loss cost determination module, configured to determine, according to the hardware device parameters and the communication transmission parameters corresponding to each of the parties, a loss cost corresponding to each of the parties by a product logarithm function;
a base loss value determining module configured to determine a base loss value corresponding to each of the participants according to the calculated loss value, the communication loss value, and the loss cost corresponding to each of the participants.
For specific definition of the federal incentive distribution device, see the above definition of the federal incentive distribution method, which is not described herein again. The various modules in the federal incentivized distribution device described above may be implemented in whole or in part in software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 10. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store data used by the federal incentive distribution method in the above embodiments. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method for federated incentive assignment.
In one embodiment, a computer device is provided, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the federal incentive distribution method in the above embodiments is implemented.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the federal incentive distribution method in the above embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method for federated incentive assignment, comprising:
receiving an initial training data set sent by each participant joining a federated system, and determining an effective training data set corresponding to each participant and a training quality vector thereof from the initial training data set; associating a total amount of valid data with a valid training data set of said participant;
determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to the participants and the total effective data amount;
determining a contribution value of each participant by adopting a marginal utility measurement method, and determining a preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant;
determining actual incentive distribution values corresponding to all the parties according to the preset incentive distribution values and a preset incentive determining strategy;
and executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
2. The federal incentive distribution method of claim 1 wherein the initial training data set includes at least one initial training data; determining, from the initial training data set, an effective training data set corresponding to each of the participants and a training quality vector corresponding to each of the effective training data sets, including:
receiving a data cleaning instruction containing a training requirement, and performing data cleaning processing on the initial training set of each participant to remove initial training data which does not meet the training requirement in the initial training set;
recording an initial training set after the initial training data which do not meet the training requirement are removed as the effective training data set;
inputting the effective training data sets into a federal feature engineering module, performing feature evaluation processing on the effective training data sets through the federal feature engineering module, and determining training quality vectors corresponding to the effective training data sets.
3. The method for federal incentive distribution according to claim 2, wherein the determining the federal incentive depth of the federal system based on the training quality vectors and the total amount of available data corresponding to each of the participants comprises:
determining a matching quality vector from the training quality vectors corresponding to each of the participants; the matching quality refers to a training quality vector with the highest matching degree with the training requirement;
determining an average quality vector by adopting a mathematical expectation algorithm according to the training quality vectors corresponding to the participants;
determining an average effective data amount by adopting a mathematical expectation algorithm according to the total effective data amount corresponding to each participant;
and acquiring the maximum data carrying capacity of the federal system, and determining the federal excitation depth according to the matching quality vector, the average effective data and the maximum data carrying capacity.
4. A federal incentive distribution method as claimed in claim 3, wherein said determining the federal incentive depth based on the matching quality vector, the mean availability data, and the maximum data carrying capacity comprises:
receiving system service parameters sent by each participant, and determining total service parameters of the federal system according to the system service parameters of each participant;
acquiring a first preset quantity decision parameter, a second preset quantity decision parameter, a first preset depth decision parameter and a second preset depth decision parameter of the federal system; the second predetermined number decision parameter is greater than the first predetermined number decision parameter;
when the average effective data volume is smaller than the first preset number decision parameter, determining the federal depth of excitation according to the total service parameters, the matching quality vectors, the average effective data and the maximum data carrying capacity;
when the average effective data volume is greater than or equal to the first preset number decision parameter and less than the second preset number decision parameter, determining the federal excitation depth according to the first preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity;
and when the average effective data volume is greater than or equal to the second preset number decision parameter, determining the federal depth of excitation according to the second preset depth decision parameter, the total service parameter, the matching quality vector, the average effective data and the maximum data carrying capacity.
5. The federal incentive distribution system of claim 1 wherein said determining the contribution of each of said participants using a marginal utility measure comprises:
determining the marginal utility of each participant on the federal system by adopting a Shapley value algorithm according to an effective training data set corresponding to each participant;
and determining a contribution value corresponding to each participant according to the marginal utility corresponding to each participant.
6. A federal incentive distribution method as defined in claim 1, wherein said determining an actual incentive distribution value corresponding to each of said participants based on said preset incentive distribution value and a preset incentive determination policy comprises:
acquiring a base loss value corresponding to each participant, and comparing the preset excitation distribution value corresponding to the same participant with the base loss value;
when the preset excitation distribution value is larger than or equal to the basic loss value, recording an actual excitation distribution value of the participant corresponding to the preset excitation distribution value as the preset excitation distribution value;
and when the preset excitation distribution value is smaller than the basic loss value, recording the actual excitation distribution value of the participant corresponding to the basic loss value as the basic loss value according to the preset excitation determination strategy.
7. The federal incentive distribution method of claim 6, wherein prior to obtaining a base loss value corresponding to each of the participants comprises:
determining a calculated loss value corresponding to each participant according to the hardware equipment parameters corresponding to each participant and the total effective data amount through a calculated loss function;
determining a communication loss value corresponding to each of the participants according to the communication transmission parameter corresponding to each of the participants through a communication loss function;
determining loss costs corresponding to each of the participants according to hardware device parameters corresponding to each of the participants and the communication transmission parameters by a product logarithm function;
determining a base loss value corresponding to each of the participants based on the calculated loss value, the communication loss value, and the loss cost corresponding to each of the participants.
8. A federated incentive dispensing device, comprising:
the system comprises a data processing module, a data processing module and a data processing module, wherein the data processing module is used for receiving an initial training data set sent by each participant who joins a federated system, and determining an effective training data set corresponding to each participant and a training quality vector thereof from the initial training data set; associating a total amount of valid data with a valid training data set of said participant;
the federal depth of excitation determination module is used for determining the federal depth of excitation of the federal system according to the training quality vectors corresponding to the participants and the total amount of effective data;
the preset incentive distribution value determining module is used for determining the contribution value of each participant by adopting a marginal utility measurement method, and determining the preset incentive distribution value corresponding to each participant according to the federal incentive depth and the contribution value corresponding to each participant;
the actual incentive distribution value determining module is used for determining an actual incentive distribution value corresponding to each party according to the preset incentive distribution value and a preset incentive determining strategy;
and the incentive distribution task execution module is used for executing the federal incentive distribution task according to the actual incentive distribution value corresponding to each participant.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the federal incentive distribution method as in any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the federal incentive distribution method as in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110449555.4A CN113159190B (en) | 2021-04-25 | 2021-04-25 | Federal incentive distribution method, apparatus, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110449555.4A CN113159190B (en) | 2021-04-25 | 2021-04-25 | Federal incentive distribution method, apparatus, computer device, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113159190A true CN113159190A (en) | 2021-07-23 |
| CN113159190B CN113159190B (en) | 2024-02-02 |
Family
ID=76870438
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110449555.4A Active CN113159190B (en) | 2021-04-25 | 2021-04-25 | Federal incentive distribution method, apparatus, computer device, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113159190B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114553869A (en) * | 2022-02-11 | 2022-05-27 | 新智我来网络科技有限公司 | Method and device for determining resource contribution degree based on joint learning and electronic equipment |
| CN116384502A (en) * | 2022-09-09 | 2023-07-04 | 京信数据科技有限公司 | Method, device, equipment and medium for calculating contribution of participant value in federal learning |
| CN116738444A (en) * | 2023-08-15 | 2023-09-12 | 山东省计算中心（国家超级计算济南中心） | Xia Puli value-based multi-party contribution degree evaluation method for data security sharing platform |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020046144A1 (en) * | 1992-10-28 | 2002-04-18 | Graff Richard A. | Further improved system and methods for computing to support decomposing property into separately valued components |
| GB2468798A (en) * | 2005-05-27 | 2010-09-22 | Ultra Scan Corp | Multimodal authorization method, system and device |
| CN112132198A (en) * | 2020-09-16 | 2020-12-25 | 建信金融科技有限责任公司 | Data processing method, device and system and server |
| US20210081966A1 (en) * | 2019-09-18 | 2021-03-18 | Divert, Inc. | Methods for sensor reuse across a supply chain |
| CN112597240A (en) * | 2021-03-01 | 2021-04-02 | 索信达(北京)数据技术有限公司 | Federal learning data processing method and system based on alliance chain |
-
2021
- 2021-04-25 CN CN202110449555.4A patent/CN113159190B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020046144A1 (en) * | 1992-10-28 | 2002-04-18 | Graff Richard A. | Further improved system and methods for computing to support decomposing property into separately valued components |
| GB2468798A (en) * | 2005-05-27 | 2010-09-22 | Ultra Scan Corp | Multimodal authorization method, system and device |
| US20210081966A1 (en) * | 2019-09-18 | 2021-03-18 | Divert, Inc. | Methods for sensor reuse across a supply chain |
| CN112132198A (en) * | 2020-09-16 | 2020-12-25 | 建信金融科技有限责任公司 | Data processing method, device and system and server |
| CN112597240A (en) * | 2021-03-01 | 2021-04-02 | 索信达(北京)数据技术有限公司 | Federal learning data processing method and system based on alliance chain |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114553869A (en) * | 2022-02-11 | 2022-05-27 | 新智我来网络科技有限公司 | Method and device for determining resource contribution degree based on joint learning and electronic equipment |
| CN116384502A (en) * | 2022-09-09 | 2023-07-04 | 京信数据科技有限公司 | Method, device, equipment and medium for calculating contribution of participant value in federal learning |
| CN116384502B (en) * | 2022-09-09 | 2024-02-20 | 京信数据科技有限公司 | Method, device, equipment and medium for calculating contribution of participant value in federal learning |
| CN116738444A (en) * | 2023-08-15 | 2023-09-12 | 山东省计算中心（国家超级计算济南中心） | Xia Puli value-based multi-party contribution degree evaluation method for data security sharing platform |
| CN116738444B (en) * | 2023-08-15 | 2023-10-31 | 山东省计算中心（国家超级计算济南中心） | Xia Puli value-based multi-party contribution degree evaluation method for data security sharing platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113159190B (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113159190A (en) | Federal incentive distribution method and device, computer equipment and storage medium | |
| US20210158216A1 (en) | Method and system for federated learning | |
| WO2014111948A1 (en) | Task assignment in crowdsourcing | |
| CN106682825A (en) | System and method for evaluating credit of Social Internet of Things based on block chain | |
| CN110231984B (en) | Multi-workflow task allocation method and device, computer equipment and storage medium | |
| CN109377218B (en) | Method, server and mobile terminal for suppressing false sensing attack | |
| CN113553377B (en) | Data sharing method and device based on block chain and federal learning | |
| CN112668877B (en) | Method and system for distributing object resource information by combining federal learning and reinforcement learning | |
| WO2021197388A1 (en) | User indexing method in federated learning and federated learning device | |
| Yuan et al. | Incentivizing federated learning under long-term energy constraint via online randomized auctions | |
| CN114781653A (en) | Model training method, system, device and storage medium based on artificial intelligence | |
| CN112036831B (en) | Human management system control method and device, readable storage medium and terminal equipment | |
| CN110766430A (en) | Resource allocation method based on machine learning algorithm | |
| CN115129481B (en) | Computing resource allocation method and device and electronic equipment | |
| CN115242800B (en) | Game theory-based mobile edge computing resource optimization method and device | |
| CN111124439A (en) | Intelligent dynamic unloading algorithm with cloud edge cooperation | |
| CN111598390B (en) | Method, device, equipment and readable storage medium for evaluating high availability of server | |
| CN116402366A (en) | Data contribution evaluation method and device based on joint learning | |
| CN113297310A (en) | Method for selecting block chain fragmentation verifier in Internet of things | |
| CN110795255B (en) | Data batch value adjusting method and device, readable storage medium and equipment | |
| CN113239879A (en) | Federal model training and certificate detection method, device, equipment and medium | |
| CN112966968A (en) | List distribution method based on artificial intelligence and related equipment | |
| CN115981876B (en) | Body-building data processing method, system and device based on cloud and fog architecture | |
| CN117094031B (en) | Industrial digital twin data privacy protection method and related medium | |
| CN117892805B (en) | Personalized federal learning method based on supernetwork and hierarchy collaborative graph aggregation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |