CN111125779A - A blockchain-based federated learning method and device - Google Patents

A blockchain-based federated learning method and device Download PDF

Info

Publication number
CN111125779A
CN111125779A CN201911304630.7A CN201911304630A CN111125779A CN 111125779 A CN111125779 A CN 111125779A CN 201911304630 A CN201911304630 A CN 201911304630A CN 111125779 A CN111125779 A CN 111125779A
Authority
CN
China
Prior art keywords
data
node
training
participant
blockchain
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.)
Pending
Application number
CN201911304630.7A
Other languages
Chinese (zh)
Inventor
孙善宝
罗清彩
金长新
徐驰
谭强
于�玲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jinan Tengming Information Technology Co ltd
Original Assignee
Shandong Inspur Artificial Intelligence Research Institute Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shandong Inspur Artificial Intelligence Research Institute Co Ltd filed Critical Shandong Inspur Artificial Intelligence Research Institute Co Ltd
Priority to CN201911304630.7A priority Critical patent/CN111125779A/en
Publication of CN111125779A publication Critical patent/CN111125779A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • Computer Security & Cryptography (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Bioethics (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请公开了一种基于区块链的联邦学习方法及装置，方法包括：确定区块链；协调者节点根据各参与者节点发送的模型原始数据，创建联邦学习任务；接收参与者节点在本地训练得到的训练数据；根据训练数据向其他参与者节点发送待更新参数，以使其他参与者节点根据待更新参数更新自身的模型参数；在模型训练完成后，根据本次训练过程中各参与者节点提供的训练数据发放奖励资源，并将奖励写入所述区块链中。相较于传统的方式，有效地解决了参与各方的互信问题；参与联邦学习的各方共同协商产生协调者节点，增加了过程的透明性；将联邦学习全程数据记录在区块链中，保证了数据操作的可追溯性；通过奖励资源鼓励各方积极参与，提升参与方的积极性。

Figure 201911304630

The present application discloses a blockchain-based federated learning method and device. The method includes: determining a blockchain; a coordinator node creates a federated learning task according to model original data sent by each participant node; The training data obtained by training; send the parameters to be updated to other participant nodes according to the training data, so that other participant nodes update their own model parameters according to the parameters to be updated; after the model training is completed, according to the The training data provided by the node issues reward resources and writes the reward into the blockchain. Compared with the traditional method, it effectively solves the problem of mutual trust among participating parties; all parties participating in federated learning negotiate to generate coordinator nodes, which increases the transparency of the process; the whole process of federated learning data is recorded in the blockchain, It ensures the traceability of data operations; encourages all parties to actively participate by rewarding resources, and enhances the enthusiasm of participants.

Figure 201911304630

Description

Block chain-based federal learning method and device
Technical Field
The application relates to the field of federal learning, in particular to a block chain-based federal learning method and a block chain-based federal learning device.
Background
Federal Learning (Federal Learning) is a new artificial intelligence basic technology, which was proposed by Google in 2016, and is originally used for solving the problem of local model updating of android mobile phone terminal users, and the design goal of the technology is to carry out efficient machine Learning among multiple parties or multiple computing nodes on the premise of guaranteeing information safety during big data exchange, protecting terminal data and personal data privacy and guaranteeing legal compliance. In most industries, data exists in an isolated island mode, and due to the problems of industry competition, privacy safety, complex administrative procedures and the like, even if data integration is realized among different departments of the same company, the important resistance is also faced, through federal learning, encryption operation is carried out on interactive data in the process of model training and reasoning, and the respective data are efficiently and accurately used together to a certain extent.
However, in the prior art, since the data contribution of each participant cannot be accurately evaluated, it is finally difficult to divide the benefit according to the contribution of each participant. And mutual trust can not be achieved among all the participants, and the training efficiency of the model during federal learning can be reduced.
Disclosure of Invention
In order to solve the above problem, the present application provides a block chain-based federal learning method, including: determining a pre-created block chain, wherein nodes of the block chain at least comprise participant nodes participating in federated learning and coordinator nodes processing data in the federated learning; the coordinator node creates a federal learning task according to the model original data sent by each participant node, and writes the federal learning task into the block chain; receiving training data obtained by local training of the participant node, and writing the training data into the block chain; sending parameters to be updated to other participant nodes according to the training data so that the other participant nodes update the model parameters of the other participant nodes according to the parameters to be updated; and after the model training is finished, issuing reward resources according to the training data provided by each participant node in the training process, and writing the reward resources into the block chain.
In one example, receiving training data obtained by the participant node training locally comprises: receiving intermediate gradient data obtained by local training of the participant nodes; sending the intermediate gradient data to a corresponding target participant node; receiving first feedback data which are returned by the target participant node and generated by training the model according to own label data; determining total gradient data in the current training round according to the first feedback data; sending parameters to be updated to other participant nodes according to the training data, wherein the parameters to be updated comprise: and sending the parameters to be updated to other participant nodes according to the total gradient data.
In one example, the issuing of reward resources according to the training data provided by each participant node in the training process includes: and issuing reward resources according to the training data provided by each participant node in the training process and the promotion effect of the training data on the model, wherein the training data comprises the intermediate gradient data and the first feedback data.
In one example, after sending the intermediate gradient data to the corresponding target participant node, the method further comprises: receiving second feedback data which are returned by the target participant node and represent that the target participant node does not own corresponding label data; and sending the intermediate gradient data to other target participant nodes.
In one example, before sending the parameters to be updated to the other participant nodes according to the training data, the method further comprises: and determining that the historical training data which is the same as the training data does not exist according to the historical training data received by the user.
In one example, the blockchain comprises a management node, an endorsement node and an accounting node, and intelligent contracts are deployed in the blockchain; the management node is used for controlling whether an external node is allowed to be deployed in the block chain as a participant node; the endorsement node is used for executing the intelligent contract and sending an execution result to the accounting node; and the accounting node is used for sending a notice to the corresponding node and processing the issuance and transaction of the reward resource.
In one example, the smart contract includes at least one of the model raw data, third party validation data, the training data, the reward resource.
In one example, the intelligent contract is for: when the original data of the model meet the preset requirements, creating the federal learning task; or the effect of the model is verified through the third party verification data, and the verification result is written into the block chain; or issuing the reward resource according to the training data.
In one example, the method further comprises: receiving inference calculation demand information sent by a first participant node; and sending the inference calculation demand information to a second participant node so that the second participant node assists the first participant node in carrying out inference calculation, and receiving the token sent by the first participant node as a reward of the transaction after the inference calculation is finished.
On the other hand, this application has still provided a block chain-based federal learning device, includes: the system comprises a determining module, a judging module and a judging module, wherein the determining module determines a pre-created block chain, and nodes of the block chain at least comprise participant nodes participating in federated learning and coordinator nodes processing data in the federated learning; the coordinator node creates a federal learning task according to the model original data sent by each participant node, and writes the federal learning task into the block chain; the receiving module is used for receiving training data obtained by local training of the participant node and writing the training data into the block chain; the updating module is used for sending parameters to be updated to other participant nodes according to the training data so that the other participant nodes update the model parameters of the other participant nodes according to the parameters to be updated; and the reward module is used for issuing reward resources according to the training data provided by each participant node in the training process after the model training is finished, and writing the reward resources into the block chain.
The federal learning method provided by the application can bring the following beneficial effects:
the block chains are effectively utilized, related interest parties participating in federal learning are combined, and block chain infrastructure is utilized to write into the block chains, so that the chain linking of the federal learning training reasoning whole life cycle is realized; compared with the traditional mode, the method effectively solves the mutual trust problem of each party through a unified and decentralized mode, increases the cooperation of each party, and reduces the trust cost and the operation cost of federal study; all parties participating in federal learning negotiate together to generate coordinator nodes, and the coordinator nodes monitor the operation process of the coordinator nodes together, so that the process transparency is improved; the data of the whole course of the federal study are recorded in the block chain, so that the traceability of data operation is ensured, related responsible parties are more effectively traced, the behaviors of federal study participants are standardized to a certain extent, all parties are encouraged to participate actively by rewarding resources, the enthusiasm of sharing high-quality data and mutual cooperation of the participants is improved, a more efficient and accurate model is formed, and the efficiency of final business is improved.
In addition, the whole process of federated learning joint modeling and reasoning can be realized in an intelligent contract form, the automatic execution of the intelligent contract also reduces the artificial influence, and the data privacy of the joint modeling process is ensured through means such as encryption transmission, digital summarization, anonymization, secure channel transmission and the like.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic flow chart of a block chain-based federal learning method in an embodiment of the present application;
FIG. 2 is a block chain-based module schematic diagram of a federated learning apparatus in an embodiment of the present application;
FIG. 3 is a block chain structure diagram according to an embodiment of the present application;
FIG. 4 is a block diagram of a blockchain according to an embodiment of the present invention;
fig. 5 is a flowchart illustrating another federal learning method based on a blockchain in the embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. 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 application.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
As shown in fig. 1 and fig. 5, an embodiment of the present application provides a block chain-based federal learning method, where the method includes:
s101, determining a pre-created block chain, wherein nodes of the block chain at least comprise participant nodes participating in federal learning and coordinator nodes processing data in the federal learning.
First, a pre-created blockchain is determined, and as shown in fig. 3, infrastructure in the blockchain may be established by a blockchain service provider of the cloud data center in a P2P (peer-to-peer) consensus network, and a corresponding consensus algorithm is selected to form a P2P consensus network.
The cloud data center gathers a large amount of computing storage resources and provides cloud infrastructure services for federal learning participants, wherein the infrastructure services comprise a trusted execution environment, an encryption and decryption service, a block chain infrastructure service and a third-party certificate authentication service provided by a third-party digital authentication center.
In particular, the trusted execution environment provides a secure execution environment of the authorization security system, protecting the confidentiality, integrity, and access rights of application resources and data running on the trusted execution environment. Particularly, when the trusted execution environment in the cloud data center is in safe connection with the hardware encryption device, exclusive and special cryptographic service is provided for the trusted execution environment. The encryption and decryption services are provided by the cloud data center through hardware encryption equipment and comprise asymmetric encryption, digital signature, identity authentication, homomorphic encryption and the like. The blockchain infrastructure services are provided by providers, blockchain infrastructure is established in a P2P consensus network, and all intelligent contract execution status results are recorded in blockchains by creating a reward resource, such as a token, to which value is assigned as a reward, and disclosing the intelligent contract programs to be executed in blockchains for all participants and community developers to review. The third-party digital certificate authentication center is responsible for issuing digital certificates, can be provided by the block chain infrastructure in a matching way, and realizes the safe communication and identity authentication between the joint modeling participants and the block chain nodes.
The block chain at least comprises two types of nodes, one type of node is a participant node participating in the federal learning, each participant node runs a training program in a local trusted execution environment, and a model is trained together through the federal learning and provides corresponding data so as to complete the federal learning task. The other type is a coordinator node, and the coordinator node processes various data in the federal learning, such as collection, calculation and distribution of relevant parameters of the model, and simultaneously issues relevant intelligent contract determination process data and evaluation results. The coordinator node may be generated and supervised jointly by the participant nodes,
s102, the coordinator node creates a federal learning task according to the model original data sent by each participant node, and writes the federal learning task into the block chain.
After determining the blockchain, the participant nodes may collectively negotiate raw data that determines the model, where the raw data may include metadata, validation data sets, and service specifications of the coordinator node. And the coordinator node initializes, operates in a trusted operation environment, is connected with the encryption and decryption service, and issues a digital certificate by a digital certificate certification center for identity certification. And then, after the coordinator node receives the original data, a federal learning task can be created, and the federal learning task is written into the block chain.
Of course, third party verification data may be provided by each participant node to the coordinator node over a secure channel after the federal learning task is written into the blockchain. And the coordinator node executes the verification data set, generates a verification result after verifying the validity, and writes the verification result into the block chain.
S103, receiving training data obtained by local training of the participant node, and writing the training data into the block chain.
And S104, sending the parameters to be updated to other participant nodes according to the training data so that the other participant nodes update the model parameters of the other participant nodes according to the parameters to be updated.
After the federal learning task is established, the participant nodes correspondingly train in respective local places according to model original data to obtain training data, and then the training data is sent to the coordinator nodes, and the coordinator nodes can write the training data into the block chain. After the training data is obtained, the parameters to be updated may be determined based on the training data. The parameters to be updated here refer to parameters that need to be updated in the model after the model is trained. And then the parameters to be updated are sent to other participant nodes, and the other participant nodes can update the parameters of the model of the self body according to the parameters to be updated.
Specifically, the training method of the federal learning task is usually trained by a gradient descent method, so that the participant node does not have corresponding label data, a complete gradient training cannot be completed during training, intermediate gradient data can be generated, and at this time, the participant node can send the encrypted intermediate gradient data to the coordinator node through a trusted channel. Then the coordinator node determines whether the same intermediate gradient data exists according to the historical training data which is received by the coordinator node and stored locally. If the intermediate gradient data exists, the intermediate gradient data is sent before, only the intermediate gradient data is recorded and stored locally, and the intermediate gradient data does not need to be forwarded. If the intermediate gradient data does not exist, the intermediate gradient data is indicated to be new data, the intermediate gradient data can be sent to a target participant node with corresponding label data, the target participant node calculates loss according to the label data owned by the target participant node, corresponding feedback data (called first feedback data) is encrypted and then returned to the coordinator node, the coordinator node determines total gradient data in the training round according to the first feedback data, and sends parameters to be updated to other participant nodes according to the total gradient data. After updating the parameters, the other participant nodes may send acknowledgement messages to the participant nodes. The participant node can determine that the gradient update of the current round is completed.
In addition, if the target participant node does not own the corresponding label data and cannot be trained through the intermediate gradient data, a second feedback data indicating that the target participant node does not own the corresponding label data can be returned, and the coordinator node sends the second feedback data to other target participant nodes.
And S105, after model training is finished, distributing reward resources according to the training data provided by each participant node in the training process, and writing the reward resources into the block chain.
The model is continuously trained by the method in the steps until the loss function is converged, the model training can be determined to be completed, at this time, reward resources can be issued to each participant node according to training data provided by each participant node in the training process, such as intermediate gradient data and first feedback data, and the reward resources are written into the block chain. The bonus resource may be a token, a priority level when writing data to the blockchain, or some authority, and will not be described herein.
In one embodiment, the blockchain further comprises a management node, an endorsement node and an accounting node, and the blockchain is also deployed with an intelligent contract. Each participant node can jointly select a management node, an endorsement node and an accounting node in the blockchain, and a third-party digital certificate authentication center issues a digital certificate to realize mutual trust among the nodes. In the election, the three nodes may be elected from the participant nodes, or a new node may be created separately, which is not limited herein. The management node is used to control whether the external node is allowed to be deployed in the blockchain as a participant node, i.e. is responsible for admission of the participant node. And the endorsement node is used for executing the intelligent contract and sending an execution result to the accounting node. The accounting nodes reach consensus, generate new blocks, complete related operations according to preset rules, send notifications to corresponding nodes, and process the issuance and transaction of reward resources.
Further, as shown in FIG. 4, a corresponding intelligent contract may be generated and deployed in a blockchain. Wherein the smart contract comprises at least one of model raw data, third party verification data, training data, and reward resources. The intelligent contract is used for: when the original data of the model meet the preset requirements, creating a federal learning task; or verifying the effect of the model through third-party verification data, and writing a verification result into the block chain; or awarding bonus resources based on the training data. Because the intelligent contract is deployed on the block chain, each node can be examined, the rights and interests of each node are guaranteed, and the convenience in data processing is improved.
In particular, intelligent contracts may include creating mission intelligent contracts that are examined after each participant node provides model raw data. And if the corresponding requirements or standards are met, creating a federal learning task and starting to train the model. The method can also comprise a model initial effect intelligent contract and a model training effect intelligent contract, each participant node provides third party verification data, and when the model is not trained, the initial effect of the model is verified and written into a block chain; and after the model training is finished, verifying the training effect of the model and writing the training effect into the block chain. The method can also comprise gradient updating intelligent contracts, and in the training process of the model, the training data sent by each participant node in the training process of the round and the corresponding promotion effect can be written into the block chain every time the training of one gradient is completed. And the intelligent reward contract is also included, and reward resources are issued according to the data volume provided by each participant node and the model improvement effect after the model training is finished.
In one embodiment, if a participant node needs other nodes to assist itself in performing inference calculations, corresponding inference calculation requirement information can be sent to the coordinator node. And then the coordinator node sends the inference calculation requirement information to the second participant node after receiving the inference calculation requirement information sent by the first participant node. And after receiving the information, the second participant node assists the first participant node to carry out reasoning calculation, and after the reasoning calculation is finished, the second participant node receives the token sent by the first participant node as a reward of the transaction. The coordinator node may then write the process of this transaction into the blockchain.
It should be noted that, in the foregoing embodiment of the present application, when data is transmitted and received and data is written into a block chain, original data may be written, or data may be hashed and then written, which is not limited herein.
The embodiment of the present application further provides a block chain-based federal learning device, including:
the determining module 201 is configured to determine a pre-created block chain, where nodes of the block chain at least include a participant node participating in federated learning and a coordinator node processing data in the federated learning;
a creating module 202, where the coordinator node creates a federal learning task according to the model raw data sent by each participant node, and writes the federal learning task into the block chain;
the receiving module 203 receives training data obtained by local training of the participant node, and writes the training data into the block chain;
the updating module 204 is used for sending parameters to be updated to other participant nodes according to the training data so that the other participant nodes update the model parameters of the other participant nodes according to the parameters to be updated;
and the reward module 205, after the model training is completed, issues reward resources according to the training data provided by each participant node in the training process, and writes the reward resources into the block chain.
The above description is merely one or more embodiments of the present disclosure and is not intended to limit the present disclosure. Various modifications and alterations to one or more embodiments of the present description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of one or more embodiments of the present specification should be included in the scope of the claims of the present specification.

Claims (10)

1.一种基于区块链的联邦学习方法，其特征在于，包括：1. A blockchain-based federated learning method, comprising: 确定预先创建的区块链，其中，所述区块链的节点中至少包括参与联邦学习的参与者节点、处理所述联邦学习中数据的协调者节点；determining a pre-created blockchain, wherein the nodes of the blockchain at least include participant nodes participating in federated learning and coordinator nodes processing data in the federated learning; 所述协调者节点根据各所述参与者节点发送的模型原始数据，创建联邦学习任务，并将所述联邦学习任务写入所述区块链；The coordinator node creates a federated learning task according to the model original data sent by each of the participant nodes, and writes the federated learning task into the blockchain; 接收所述参与者节点在本地训练得到的训练数据，并将所述训练数据写入所述区块链中；Receive the training data obtained by the local training of the participant node, and write the training data into the blockchain; 根据所述训练数据向其他所述参与者节点发送待更新参数，以使所述其他参与者节点根据所述待更新参数更新自身的模型参数；Send the parameters to be updated to the other participant nodes according to the training data, so that the other participant nodes update their own model parameters according to the parameters to be updated; 在模型训练完成后，根据本次训练过程中各参与者节点提供的所述训练数据发放奖励资源，并将所述奖励资源写入所述区块链中。After the model training is completed, reward resources are issued according to the training data provided by each participant node in the training process, and the reward resources are written into the blockchain. 2.根据权利要求1所述的方法，其特征在于，接收所述参与者节点在本地训练得到的训练数据，包括：2. The method according to claim 1, wherein receiving the training data obtained by the local training of the participant node comprises: 接收所述参与者节点在本地训练得到的中间梯度数据；receiving the intermediate gradient data obtained by the local training of the participant node; 将所述中间梯度数据发送至对应的目标参与者节点处；sending the intermediate gradient data to the corresponding target participant node; 接收所述目标参与者节点返回的，根据自身所拥有的标签数据训练所述模型生成的第一反馈数据；Receive the first feedback data generated by training the model according to the label data owned by the target participant node; 根据所述第一反馈数据确定在本轮训练中的总梯度数据；Determine the total gradient data in this round of training according to the first feedback data; 根据所述训练数据向其他所述参与者节点发送待更新参数，包括：Send the parameters to be updated to the other participant nodes according to the training data, including: 根据所述总梯度数据向其他所述参与者节点发送待更新参数。The parameters to be updated are sent to the other participant nodes according to the total gradient data. 3.根据权利要求2所述的方法，其特征在于，根据本次训练过程中各参与者节点提供的所述训练数据发放奖励资源，包括：3. The method according to claim 2, wherein the distribution of reward resources according to the training data provided by each participant node in this training process, comprises: 根据本次训练过程中各参与者节点提供的所述训练数据，以及所述训练数据对模型的提升效果发放奖励资源，其中，所述训练数据包括所述中间梯度数据和所述第一反馈数据。Reward resources are distributed according to the training data provided by each participant node in this training process and the improvement effect of the training data on the model, wherein the training data includes the intermediate gradient data and the first feedback data . 4.根据权利要求2所述的方法，其特征在于，将所述中间梯度数据发送至对应的目标参与者节点处之后，所述方法还包括：4. The method according to claim 2, wherein after the intermediate gradient data is sent to the corresponding target participant node, the method further comprises: 接收所述目标参与者节点返回的，表示自身未拥有相应的标签数据的第二反馈数据；Receive the second feedback data returned by the target participant node, indicating that it does not own the corresponding label data; 将所述中间梯度数据发送至其他的目标参与者节点处。Send the intermediate gradient data to other target participant nodes. 5.根据权利要求1所述的方法，其特征在于，根据所述训练数据向其他所述参与者节点发送待更新参数之前，所述方法还包括：5. The method according to claim 1, wherein before sending the parameters to be updated to other said participant nodes according to the training data, the method further comprises: 根据自身所收到的历史训练数据，确定不存在与所述训练数据相同的所述历史训练数据。According to the historical training data received by itself, it is determined that the historical training data identical to the training data does not exist. 6.根据权利要求1所述的方法，其特征在于，所述区块链中包括管理节点、背书节点和记账节点，并且所述区块链中部署有智能合约；6. The method according to claim 1, wherein the blockchain includes a management node, an endorsement node and an accounting node, and a smart contract is deployed in the blockchain; 所述管理节点，用于控制外部节点是否允许作为参与者节点部署在所述区块链中；the management node, for controlling whether an external node is allowed to be deployed in the blockchain as a participant node; 所述背书节点，用于执行所述智能合约，并将执行结果发送至所述记账节点；The endorsement node is used to execute the smart contract and send the execution result to the accounting node; 所述记账节点，用于向相应的节点发送通知，并处理奖励资源的发放和交易。The accounting node is used to send a notification to the corresponding node and process the issuance and transaction of reward resources. 7.根据权利要求6所述的方法，其特征在于，所述智能合约包括所述模型原始数据、第三方验证数据、所述训练数据、所述奖励资源中的至少一种。7. The method according to claim 6, wherein the smart contract includes at least one of the model original data, third-party verification data, the training data, and the reward resource. 8.根据权利要求7所述的方法，其特征在于，所述智能合约用于：8. The method of claim 7, wherein the smart contract is used to: 在所述模型原始数据符合预设要求时，创建所述联邦学习任务；或Create the federated learning task when the model raw data meets preset requirements; or 通过所述第三方验证数据对所述模型的效果进行验证，并将验证结果写入所述区块链中；或Verify the effect of the model through the third-party verification data, and write the verification result into the blockchain; or 根据所述训练数据发放所述奖励资源。The reward resource is issued according to the training data. 9.根据权利要求6所述的方法，其特征在于，所述方法还包括：9. The method according to claim 6, wherein the method further comprises: 接收第一参与者节点发送的推理计算需求信息；Receive the inference calculation requirement information sent by the first participant node; 向第二参与者节点发送所述推理计算需求信息，以使所述第二参与者节点辅助所述第一参与者节点进行推理计算，并在推理计算结束后，接收所述第一参与者节点发送的代币作为本次交易的报酬。Send the inference calculation requirement information to the second participant node, so that the second participant node assists the first participant node to perform inference calculation, and after the inference calculation is completed, receive the first participant node The tokens sent are the reward for this transaction. 10.一种基于区块链的联邦学习装置，其特征在于，包括：10. A blockchain-based federated learning device, comprising: 确定模块，确定预先创建的区块链，其中，所述区块链的节点中至少包括参与联邦学习的参与者节点、处理所述联邦学习中数据的协调者节点；a determination module to determine a pre-created blockchain, wherein the nodes of the blockchain at least include participant nodes participating in federated learning and coordinator nodes processing data in the federated learning; 创建模块，所述协调者节点根据各所述参与者节点发送的模型原始数据，创建联邦学习任务，并将所述联邦学习任务写入所述区块链；A creation module, wherein the coordinator node creates a federated learning task according to the model original data sent by each of the participant nodes, and writes the federated learning task into the blockchain; 接收模块，接收所述参与者节点在本地训练得到的训练数据，并将所述训练数据写入所述区块链中；a receiving module, receiving the training data obtained by the local training of the participant node, and writing the training data into the blockchain; 更新模块，根据所述训练数据向其他所述参与者节点发送待更新参数，以使所述其他参与者节点根据所述待更新参数更新自身的模型参数；an update module, sending parameters to be updated to other said participant nodes according to said training data, so that said other participant nodes update their own model parameters according to said to-be-updated parameters; 奖励模块，在模型训练完成后，根据本次训练过程中各参与者节点提供的所述训练数据发放奖励资源，并将所述奖励资源写入所述区块链中。The reward module, after the model training is completed, distributes reward resources according to the training data provided by each participant node in the training process, and writes the reward resources into the blockchain.
CN201911304630.7A 2019-12-17 2019-12-17 A blockchain-based federated learning method and device Pending CN111125779A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911304630.7A CN111125779A (en) 2019-12-17 2019-12-17 A blockchain-based federated learning method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911304630.7A CN111125779A (en) 2019-12-17 2019-12-17 A blockchain-based federated learning method and device

Publications (1)

Publication Number Publication Date
CN111125779A true CN111125779A (en) 2020-05-08

Family

ID=70498324

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911304630.7A Pending CN111125779A (en) 2019-12-17 2019-12-17 A blockchain-based federated learning method and device

Country Status (1)

Country Link
CN (1) CN111125779A (en)

Cited By (68)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111340242A (en) * 2020-05-15 2020-06-26 支付宝(杭州)信息技术有限公司 Model joint training method and device for protecting privacy
CN111539731A (en) * 2020-06-19 2020-08-14 支付宝(杭州)信息技术有限公司 Block chain-based federal learning method and device and electronic equipment
CN111614679A (en) * 2020-05-22 2020-09-01 深圳前海微众银行股份有限公司 Federated learning qualification recovery method, device and readable storage medium
CN111612168A (en) * 2020-06-30 2020-09-01 腾讯科技（深圳）有限公司 Management method and related device for machine learning task
CN111652382A (en) * 2020-06-04 2020-09-11 深圳前海微众银行股份有限公司 Blockchain-based data processing method, device, equipment and storage medium
CN111698322A (en) * 2020-06-11 2020-09-22 福州数据技术研究院有限公司 Medical data safety sharing method based on block chain and federal learning
CN111723946A (en) * 2020-06-19 2020-09-29 深圳前海微众银行股份有限公司 A federated learning method and device applied to blockchain
CN111858753A (en) * 2020-06-30 2020-10-30 全链通有限公司 Block chain-based training parameter processing method, device and storage medium
CN111901309A (en) * 2020-07-03 2020-11-06 北京邮电大学 A data security sharing method, system and device
CN111930698A (en) * 2020-07-01 2020-11-13 南京晓庄学院 Data security sharing method based on Hash diagram and federal learning
CN111931242A (en) * 2020-09-30 2020-11-13 国网浙江省电力有限公司电力科学研究院 Data sharing method, computer equipment applying same and readable storage medium
CN111950739A (en) * 2020-08-13 2020-11-17 深圳前海微众银行股份有限公司 Blockchain-based data processing method, device, equipment and medium
CN111967610A (en) * 2020-08-18 2020-11-20 中国银行股份有限公司 Block chain-based federal learning incentive method, device, equipment and storage medium
CN111966698A (en) * 2020-07-03 2020-11-20 华南师范大学 Credible federal learning method, system, device and medium based on block chain
CN111985000A (en) * 2020-08-21 2020-11-24 深圳前海微众银行股份有限公司 Model service output method, device, equipment and storage medium
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
CN112182102A (en) * 2020-09-23 2021-01-05 西安纸贵互联网科技有限公司 Data processing method, device, electronic device and storage medium in federated learning
CN112232437A (en) * 2020-11-04 2021-01-15 深圳技术大学 Internet of things terminal data analysis method and system
CN112232527A (en) * 2020-09-21 2021-01-15 北京邮电大学 Safe distributed federal deep learning method
CN112348204A (en) * 2020-11-05 2021-02-09 大连理工大学 A method for data security sharing of marine Internet of things under the framework of edge computing based on federated learning and blockchain technology
CN112434818A (en) * 2020-11-19 2021-03-02 脸萌有限公司 Model construction method, device, medium and electronic equipment
CN112434313A (en) * 2020-11-11 2021-03-02 北京邮电大学 Data sharing method, system, electronic device and storage medium
CN112465117A (en) * 2020-11-25 2021-03-09 平安科技（深圳）有限公司 Contract generation model construction method, device, equipment and storage medium
CN112540926A (en) * 2020-12-17 2021-03-23 杭州趣链科技有限公司 Resource allocation fairness federal learning method based on block chain
CN112598132A (en) * 2020-12-04 2021-04-02 光大科技有限公司 Model training method and device, storage medium and electronic device
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
CN112634027A (en) * 2020-12-30 2021-04-09 杭州趣链科技有限公司 Self-adaptive federal parameter aggregation method for credit assessment of small and micro enterprises
CN112651592A (en) * 2020-11-27 2021-04-13 科技谷（厦门）信息技术有限公司 Enterprise credit assessment system based on multimodal transport
CN112671613A (en) * 2020-12-28 2021-04-16 深圳市彬讯科技有限公司 Federal learning cluster monitoring method, device, equipment and medium
CN112668049A (en) * 2020-12-30 2021-04-16 杭州趣链科技有限公司 Alliance chain processing method for multi-party joint calculation
CN112765677A (en) * 2020-12-30 2021-05-07 杭州溪塔科技有限公司 Block chain-based federal learning method, device and system
CN112784994A (en) * 2020-12-31 2021-05-11 浙江大学 Block chain-based federated learning data participant contribution value calculation and excitation method
CN112801307A (en) * 2021-04-13 2021-05-14 深圳索信达数据技术有限公司 Block chain-based federal learning method and device and computer equipment
CN112861152A (en) * 2021-02-08 2021-05-28 北京航空航天大学 Federal learning incentive method and system based on permit chain
CN113037722A (en) * 2021-02-26 2021-06-25 济南浪潮高新科技投资发展有限公司 Intrusion detection method and device for edge calculation scene
CN113194126A (en) * 2021-04-21 2021-07-30 泉州华中科技大学智能制造研究院 Block chain-based transverse federated learning model construction method
CN113191090A (en) * 2021-05-31 2021-07-30 中国银行股份有限公司 Block chain-based federal modeling method and device
CN113204795A (en) * 2021-07-07 2021-08-03 数据堂（北京）科技股份有限公司 Data sharing method and device based on federal data and federal service
CN113268758A (en) * 2021-06-17 2021-08-17 上海万向区块链股份公司 Data sharing system, method, medium and device based on federal learning
CN113283990A (en) * 2021-06-03 2021-08-20 光大科技有限公司 Data sharing processing method and device
CN113408746A (en) * 2021-06-22 2021-09-17 深圳大学 Block chain-based distributed federal learning method and device and terminal equipment
CN113420335A (en) * 2021-08-24 2021-09-21 浙江数秦科技有限公司 Block chain-based federal learning system
CN113468133A (en) * 2021-05-23 2021-10-01 杭州医康慧联科技股份有限公司 Online sharing system suitable for data model
CN113467927A (en) * 2021-05-20 2021-10-01 杭州趣链科技有限公司 Block chain based trusted participant federated learning method and device
CN113542220A (en) * 2021-06-09 2021-10-22 浙江泰科数联信息技术有限公司 Data security sharing method based on block chain
CN113643390A (en) * 2021-10-13 2021-11-12 启东市永圣医疗用品厂 Medical image compression storage method based on block chain and federal learning
JP2021193568A (en) * 2020-10-16 2021-12-23 ベイジン バイドゥ ネットコム サイエンス テクノロジー カンパニー リミテッド Federation learning method and device for improving matching efficiency, electronic device, and medium
CN113868675A (en) * 2020-06-30 2021-12-31 杭州海康威视数字技术股份有限公司 A deep learning data sharing method and device based on blockchain technology
CN113987077A (en) * 2021-12-23 2022-01-28 太极计算机股份有限公司 Data sensing and cross-link scheduling method and device based on chain code mechanism
CN114004265A (en) * 2020-07-13 2022-02-01 华为技术有限公司 A model training method and node device
CN114118438A (en) * 2021-10-18 2022-03-01 华北电力大学 A privacy-preserving machine learning training and reasoning method and system based on blockchain
CN114139722A (en) * 2021-11-29 2022-03-04 广发银行股份有限公司 Blockchain-based Federated Learning Task Scheduling Method, System, Equipment and Medium
CN114187006A (en) * 2021-11-03 2022-03-15 杭州未名信科科技有限公司 Block chain supervision-based federal learning method
CN114358907A (en) * 2022-01-06 2022-04-15 广东启链科技有限公司 Financial risk prediction method and system based on block chain federal learning
CN114417421A (en) * 2022-01-26 2022-04-29 深圳技术大学 A metaverse-based shared information privacy protection method and related device
CN114443754A (en) * 2020-11-03 2022-05-06 中国电信股份有限公司 Blockchain-based federated learning processing method, device, system and medium
CN114548473A (en) * 2020-11-26 2022-05-27 新智数字科技有限公司 Resource allocation method, device, readable medium and electronic equipment
CN114548472A (en) * 2020-11-26 2022-05-27 新智数字科技有限公司 Resource allocation method, device, readable medium and electronic equipment
CN114580665A (en) * 2022-03-15 2022-06-03 星环信息科技(上海)股份有限公司 Federated learning system, method, device, equipment and storage medium
CN114580658A (en) * 2021-12-28 2022-06-03 天翼云科技有限公司 Blockchain-based federated learning incentive method, device, equipment and medium
CN114638449A (en) * 2020-12-16 2022-06-17 新智数字科技有限公司 Electricity selling scheme determining method and device, readable medium and electronic equipment
CN114723441A (en) * 2021-01-05 2022-07-08 中国移动通信有限公司研究院 Method, device and equipment for constraining behaviors of demander and participator
CN114764707A (en) * 2021-01-04 2022-07-19 中国移动通信有限公司研究院 Federal learning model training method and system
CN114764475A (en) * 2021-01-14 2022-07-19 新智数字科技有限公司 Central calibration system of combined learning engine
CN114819197A (en) * 2022-06-27 2022-07-29 杭州同花顺数据开发有限公司 Block chain alliance-based federal learning method, system, device and storage medium
CN115345317A (en) * 2022-08-05 2022-11-15 北京交通大学 Fair reward distribution method based on fairness theory and oriented to federal learning
WO2023141809A1 (en) * 2022-01-26 2023-08-03 深圳技术大学 Metaverse-based shared information privacy protection method and related apparatus
CN116961939A (en) * 2022-04-14 2023-10-27 苏州科技大学 Trusted federation learning method based on alliance chain

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110443063A (en) * 2019-06-26 2019-11-12 电子科技大学 The method of the federal deep learning of self adaptive protection privacy
CN110490738A (en) * 2019-08-06 2019-11-22 深圳前海微众银行股份有限公司 A kind of federal learning method of mixing and framework
CN110490330A (en) * 2019-08-16 2019-11-22 安徽航天信息有限公司 A kind of distributed machines learning system based on block chain
CN110490335A (en) * 2019-08-07 2019-11-22 深圳前海微众银行股份有限公司 A kind of method and device calculating participant's contribution rate

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110443063A (en) * 2019-06-26 2019-11-12 电子科技大学 The method of the federal deep learning of self adaptive protection privacy
CN110490738A (en) * 2019-08-06 2019-11-22 深圳前海微众银行股份有限公司 A kind of federal learning method of mixing and framework
CN110490335A (en) * 2019-08-07 2019-11-22 深圳前海微众银行股份有限公司 A kind of method and device calculating participant's contribution rate
CN110490330A (en) * 2019-08-16 2019-11-22 安徽航天信息有限公司 A kind of distributed machines learning system based on block chain

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王亚珅: "面向数据共享交换的联邦学习技术发展综述", 《无人系统技术》 *

Cited By (103)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111340242B (en) * 2020-05-15 2020-08-04 支付宝(杭州)信息技术有限公司 Model joint training method and device for protecting privacy
CN111340242A (en) * 2020-05-15 2020-06-26 支付宝(杭州)信息技术有限公司 Model joint training method and device for protecting privacy
CN111614679A (en) * 2020-05-22 2020-09-01 深圳前海微众银行股份有限公司 Federated learning qualification recovery method, device and readable storage medium
CN111652382A (en) * 2020-06-04 2020-09-11 深圳前海微众银行股份有限公司 Blockchain-based data processing method, device, equipment and storage medium
CN111652382B (en) * 2020-06-04 2021-08-06 深圳前海微众银行股份有限公司 Blockchain-based data processing method, device, equipment and storage medium
CN111698322A (en) * 2020-06-11 2020-09-22 福州数据技术研究院有限公司 Medical data safety sharing method based on block chain and federal learning
CN111723946A (en) * 2020-06-19 2020-09-29 深圳前海微众银行股份有限公司 A federated learning method and device applied to blockchain
CN111539731A (en) * 2020-06-19 2020-08-14 支付宝(杭州)信息技术有限公司 Block chain-based federal learning method and device and electronic equipment
CN113868675A (en) * 2020-06-30 2021-12-31 杭州海康威视数字技术股份有限公司 A deep learning data sharing method and device based on blockchain technology
CN111858753A (en) * 2020-06-30 2020-10-30 全链通有限公司 Block chain-based training parameter processing method, device and storage medium
CN111612168A (en) * 2020-06-30 2020-09-01 腾讯科技（深圳）有限公司 Management method and related device for machine learning task
CN111858753B (en) * 2020-06-30 2023-11-17 全链通有限公司 Training parameter processing method, device and storage medium based on block chain
CN111930698B (en) * 2020-07-01 2024-03-15 南京晓庄学院 Data security sharing method based on hash map and federal learning
CN111930698A (en) * 2020-07-01 2020-11-13 南京晓庄学院 Data security sharing method based on Hash diagram and federal learning
CN111901309A (en) * 2020-07-03 2020-11-06 北京邮电大学 A data security sharing method, system and device
CN111966698B (en) * 2020-07-03 2023-06-13 华南师范大学 A blockchain-based trusted federated learning method, system, device and medium
CN111901309B (en) * 2020-07-03 2022-03-22 北京邮电大学 Data security sharing method, system and device
CN111966698A (en) * 2020-07-03 2020-11-20 华南师范大学 Credible federal learning method, system, device and medium based on block chain
CN114004265A (en) * 2020-07-13 2022-02-01 华为技术有限公司 A model training method and node device
CN111950739A (en) * 2020-08-13 2020-11-17 深圳前海微众银行股份有限公司 Blockchain-based data processing method, device, equipment and medium
CN111967610A (en) * 2020-08-18 2020-11-20 中国银行股份有限公司 Block chain-based federal learning incentive method, device, equipment and storage medium
CN111967610B (en) * 2020-08-18 2024-02-23 中国银行股份有限公司 Federal learning incentive method, device, equipment and storage medium based on block chain
CN111985000A (en) * 2020-08-21 2020-11-24 深圳前海微众银行股份有限公司 Model service output method, device, equipment and storage medium
CN112132198B (en) * 2020-09-16 2021-06-04 建信金融科技有限责任公司 Data processing method, device and system and server
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
CN112232527B (en) * 2020-09-21 2024-01-23 北京邮电大学 Safe distributed federal deep learning method
CN112232527A (en) * 2020-09-21 2021-01-15 北京邮电大学 Safe distributed federal deep learning method
CN112182102A (en) * 2020-09-23 2021-01-05 西安纸贵互联网科技有限公司 Data processing method, device, electronic device and storage medium in federated learning
CN111931242A (en) * 2020-09-30 2020-11-13 国网浙江省电力有限公司电力科学研究院 Data sharing method, computer equipment applying same and readable storage medium
EP3859557A3 (en) * 2020-10-16 2021-12-29 Beijing Baidu Netcom Science And Technology Co. Ltd. Federated learning method and device for improving matching efficiency, electronic device, and medium
JP2021193568A (en) * 2020-10-16 2021-12-23 ベイジン バイドゥ ネットコム サイエンス テクノロジー カンパニー リミテッド Federation learning method and device for improving matching efficiency, electronic device, and medium
CN114443754A (en) * 2020-11-03 2022-05-06 中国电信股份有限公司 Blockchain-based federated learning processing method, device, system and medium
CN112232437A (en) * 2020-11-04 2021-01-15 深圳技术大学 Internet of things terminal data analysis method and system
CN112348204A (en) * 2020-11-05 2021-02-09 大连理工大学 A method for data security sharing of marine Internet of things under the framework of edge computing based on federated learning and blockchain technology
CN112434313A (en) * 2020-11-11 2021-03-02 北京邮电大学 Data sharing method, system, electronic device and storage medium
CN112434313B (en) * 2020-11-11 2024-12-10 北京邮电大学 Data sharing method, system, electronic device and storage medium
CN112434818B (en) * 2020-11-19 2023-09-26 脸萌有限公司 Model construction method, device, medium and electronic equipment
CN112434818A (en) * 2020-11-19 2021-03-02 脸萌有限公司 Model construction method, device, medium and electronic equipment
CN112465117A (en) * 2020-11-25 2021-03-09 平安科技（深圳）有限公司 Contract generation model construction method, device, equipment and storage medium
CN112465117B (en) * 2020-11-25 2024-05-07 平安科技（深圳）有限公司 Contract generation model construction method, device, equipment and storage medium
CN114548473A (en) * 2020-11-26 2022-05-27 新智数字科技有限公司 Resource allocation method, device, readable medium and electronic equipment
CN114548472A (en) * 2020-11-26 2022-05-27 新智数字科技有限公司 Resource allocation method, device, readable medium and electronic equipment
CN112651592B (en) * 2020-11-27 2022-05-06 科技谷（厦门）信息技术有限公司 Enterprise credit assessment system based on multimodal transport
CN112651592A (en) * 2020-11-27 2021-04-13 科技谷（厦门）信息技术有限公司 Enterprise credit assessment system based on multimodal transport
CN112598132A (en) * 2020-12-04 2021-04-02 光大科技有限公司 Model training method and device, storage medium and electronic device
CN114638449A (en) * 2020-12-16 2022-06-17 新智数字科技有限公司 Electricity selling scheme determining method and device, readable medium and electronic equipment
CN112540926B (en) * 2020-12-17 2024-10-15 杭州趣链科技有限公司 Federal learning method for fair resource allocation based on blockchain
CN112540926A (en) * 2020-12-17 2021-03-23 杭州趣链科技有限公司 Resource allocation fairness federal learning method based on block chain
CN112671613B (en) * 2020-12-28 2022-08-23 深圳市彬讯科技有限公司 Federal learning cluster monitoring method, device, equipment and medium
CN112671613A (en) * 2020-12-28 2021-04-16 深圳市彬讯科技有限公司 Federal learning cluster monitoring method, device, equipment and medium
CN112634027A (en) * 2020-12-30 2021-04-09 杭州趣链科技有限公司 Self-adaptive federal parameter aggregation method for credit assessment of small and micro enterprises
CN112668049A (en) * 2020-12-30 2021-04-16 杭州趣链科技有限公司 Alliance chain processing method for multi-party joint calculation
CN112765677A (en) * 2020-12-30 2021-05-07 杭州溪塔科技有限公司 Block chain-based federal learning method, device and system
CN112765677B (en) * 2020-12-30 2024-01-23 杭州溪塔科技有限公司 Federal learning method, device and system based on blockchain
CN112784994B (en) * 2020-12-31 2023-03-03 浙江大学 Blockchain-based federated learning data participant contribution value calculation and incentive method
CN112784994A (en) * 2020-12-31 2021-05-11 浙江大学 Block chain-based federated learning data participant contribution value calculation and excitation method
CN114764707A (en) * 2021-01-04 2022-07-19 中国移动通信有限公司研究院 Federal learning model training method and system
CN114764707B (en) * 2021-01-04 2025-04-04 中国移动通信有限公司研究院 Federated learning model training method and system
CN114723441A (en) * 2021-01-05 2022-07-08 中国移动通信有限公司研究院 Method, device and equipment for constraining behaviors of demander and participator
CN114723441B (en) * 2021-01-05 2025-04-15 中国移动通信有限公司研究院 A method, device and equipment for constraining the behavior of demanders and participants
CN114764475A (en) * 2021-01-14 2022-07-19 新智数字科技有限公司 Central calibration system of combined learning engine
CN112861152B (en) * 2021-02-08 2024-02-20 北京航空航天大学 A federated learning incentive method and system based on permission chain
CN112861152A (en) * 2021-02-08 2021-05-28 北京航空航天大学 Federal learning incentive method and system based on permit chain
CN113037722B (en) * 2021-02-26 2022-06-07 山东浪潮科学研究院有限公司 Intrusion detection method and device for edge calculation scene
CN113037722A (en) * 2021-02-26 2021-06-25 济南浪潮高新科技投资发展有限公司 Intrusion detection method and device for edge calculation scene
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
CN112597240B (en) * 2021-03-01 2021-06-04 索信达(北京)数据技术有限公司 Federal learning data processing method and system based on alliance chain
CN112801307A (en) * 2021-04-13 2021-05-14 深圳索信达数据技术有限公司 Block chain-based federal learning method and device and computer equipment
CN113194126B (en) * 2021-04-21 2023-10-27 泉州华中科技大学智能制造研究院 Transverse federal learning model construction method based on blockchain
CN113194126A (en) * 2021-04-21 2021-07-30 泉州华中科技大学智能制造研究院 Block chain-based transverse federated learning model construction method
CN113467927B (en) * 2021-05-20 2024-12-27 杭州趣链科技有限公司 A blockchain-based federated learning method and device that can be trusted by participants
CN113467927A (en) * 2021-05-20 2021-10-01 杭州趣链科技有限公司 Block chain based trusted participant federated learning method and device
CN113468133A (en) * 2021-05-23 2021-10-01 杭州医康慧联科技股份有限公司 Online sharing system suitable for data model
CN113191090A (en) * 2021-05-31 2021-07-30 中国银行股份有限公司 Block chain-based federal modeling method and device
CN113283990A (en) * 2021-06-03 2021-08-20 光大科技有限公司 Data sharing processing method and device
CN113283990B (en) * 2021-06-03 2024-02-09 光大科技有限公司 Data sharing processing method and device
CN113542220A (en) * 2021-06-09 2021-10-22 浙江泰科数联信息技术有限公司 Data security sharing method based on block chain
CN113268758A (en) * 2021-06-17 2021-08-17 上海万向区块链股份公司 Data sharing system, method, medium and device based on federal learning
CN113408746A (en) * 2021-06-22 2021-09-17 深圳大学 Block chain-based distributed federal learning method and device and terminal equipment
CN113408746B (en) * 2021-06-22 2023-03-14 深圳大学 Distributed federal learning method and device based on block chain and terminal equipment
CN113204795A (en) * 2021-07-07 2021-08-03 数据堂（北京）科技股份有限公司 Data sharing method and device based on federal data and federal service
CN113420335B (en) * 2021-08-24 2021-11-12 浙江数秦科技有限公司 A blockchain-based federated learning system
CN113420335A (en) * 2021-08-24 2021-09-21 浙江数秦科技有限公司 Block chain-based federal learning system
CN113643390B (en) * 2021-10-13 2022-02-18 启东市永圣医疗用品厂 Medical image compression storage method based on block chain and federal learning
CN113643390A (en) * 2021-10-13 2021-11-12 启东市永圣医疗用品厂 Medical image compression storage method based on block chain and federal learning
CN114118438B (en) * 2021-10-18 2023-07-21 华北电力大学 A blockchain-based privacy-preserving machine learning training and reasoning method and system
CN114118438A (en) * 2021-10-18 2022-03-01 华北电力大学 A privacy-preserving machine learning training and reasoning method and system based on blockchain
CN114187006A (en) * 2021-11-03 2022-03-15 杭州未名信科科技有限公司 Block chain supervision-based federal learning method
CN114139722A (en) * 2021-11-29 2022-03-04 广发银行股份有限公司 Blockchain-based Federated Learning Task Scheduling Method, System, Equipment and Medium
CN113987077A (en) * 2021-12-23 2022-01-28 太极计算机股份有限公司 Data sensing and cross-link scheduling method and device based on chain code mechanism
CN113987077B (en) * 2021-12-23 2022-03-29 太极计算机股份有限公司 Data sensing and cross-link scheduling method and device based on chain code mechanism
CN114580658A (en) * 2021-12-28 2022-06-03 天翼云科技有限公司 Blockchain-based federated learning incentive method, device, equipment and medium
CN114358907A (en) * 2022-01-06 2022-04-15 广东启链科技有限公司 Financial risk prediction method and system based on block chain federal learning
WO2023141809A1 (en) * 2022-01-26 2023-08-03 深圳技术大学 Metaverse-based shared information privacy protection method and related apparatus
CN114417421A (en) * 2022-01-26 2022-04-29 深圳技术大学 A metaverse-based shared information privacy protection method and related device
CN114417421B (en) * 2022-01-26 2024-08-20 深圳技术大学 Shared information privacy protection method and related device based on meta universe
CN114580665B (en) * 2022-03-15 2023-10-20 星环信息科技(上海)股份有限公司 Federal learning system, method, device, equipment and storage medium
CN114580665A (en) * 2022-03-15 2022-06-03 星环信息科技(上海)股份有限公司 Federated learning system, method, device, equipment and storage medium
CN116961939B (en) * 2022-04-14 2024-07-19 苏州科技大学 Trusted federation learning method based on alliance chain
CN116961939A (en) * 2022-04-14 2023-10-27 苏州科技大学 Trusted federation learning method based on alliance chain
CN114819197A (en) * 2022-06-27 2022-07-29 杭州同花顺数据开发有限公司 Block chain alliance-based federal learning method, system, device and storage medium
CN114819197B (en) * 2022-06-27 2023-07-04 杭州同花顺数据开发有限公司 Federal learning method, system, device and storage medium based on blockchain alliance
CN115345317A (en) * 2022-08-05 2022-11-15 北京交通大学 Fair reward distribution method based on fairness theory and oriented to federal learning

Similar Documents

Publication Publication Date Title
CN111125779A (en) A blockchain-based federated learning method and device
Yue et al. A survey of decentralizing applications via blockchain: The 5G and beyond perspective
Lu et al. Zebralancer: Private and anonymous crowdsourcing system atop open blockchain
Du et al. An optimized consortium blockchain for medical information sharing
CN109450638B (en) Block chain-based electronic component data management system and method
US11157833B2 (en) Learning service blockchain
CN108256864B (en) Foundation across chain alliance and communication means, system between a kind of block chain
JP2023134800A (en) Smart contract execution using distributed coordination
CN113992360B (en) Federal learning method and equipment based on block chain crossing
CN109167695A (en) Alliance Network construction method, equipment and readable storage medium storing program for executing based on federation's study
CN111461712B (en) Transaction privacy protection and hierarchical supervision in blockchain supply chain financial scenarios
Huang et al. Blockchain-assisted personalized car insurance with privacy preservation and fraud resistance
CN110275891A (en) Artificial intelligence software market
Hu et al. Reputation-based distributed knowledge sharing system in blockchain
CN112016954A (en) Resource allocation method and device based on block chain network technology and electronic equipment
Zhang et al. SABlockFL: a blockchain-based smart agent system architecture and its application in federated learning
Baniata et al. Prifob: a privacy-aware fog-enhanced blockchain-based system for global accreditation and credential verification
Wen et al. Application of blockchain technology in data management: advantages and solutions
Keshavarzkalhori et al. Federify: A verifiable federated learning scheme based on zksnarks and blockchain
Reshi et al. The blockchain conundrum: An in‐depth examination of challenges, contributing technologies, and alternatives
CN118214761A (en) Attribute-based encrypted data sharing method, system and device based on blockchain
Zhu et al. The governance technology for blockchain systems: a survey
Yu et al. Blockchain-based solutions for mobile crowdsensing: A comprehensive survey
Scaffino et al. Alba: The Dawn of Scalable Bridges for Blockchains
Dong et al. Blockchain-Assisted Reputation Mechanism for Distributed Cloud Storage

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20210716

Address after: 250101 building S02, 1036 Chaochao Road, high tech Zone, Jinan City, Shandong Province

Applicant after: Shandong Inspur Scientific Research Institute Co.,Ltd.

Address before: Floor 6, Chaochao Road, Shandong Province

Applicant before: SHANDONG INSPUR ARTIFICIAL INTELLIGENCE RESEARCH INSTITUTE Co.,Ltd.

TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20210820

Address after: 250014 North 6th floor, S05 building, Inspur Science Park, No. 1036, Inspur Road, high tech Zone, Jinan City, Shandong Province

Applicant after: Jinan Tengming Information Technology Co.,Ltd.

Address before: 250101 building S02, 1036 Chaochao Road, high tech Zone, Jinan City, Shandong Province

Applicant before: Shandong Inspur Scientific Research Institute Co.,Ltd.

RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20200508