CN111212110A - Block chain-based federal learning system and method - Google Patents
Block chain-based federal learning system and method Download PDFInfo
- Publication number
- CN111212110A CN111212110A CN201911285920.1A CN201911285920A CN111212110A CN 111212110 A CN111212110 A CN 111212110A CN 201911285920 A CN201911285920 A CN 201911285920A CN 111212110 A CN111212110 A CN 111212110A
- Authority
- CN
- China
- Prior art keywords
- training
- node
- information
- model
- block chain
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/82—Miscellaneous aspects
- H04L47/827—Aggregation of resource allocation or reservation requests
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a block chain-based federal learning system and a block chain-based federal learning method, wherein the system comprises the following steps: the model training module is used for updating a machine learning model in the federal learning process and aggregating the change values of the machine learning model; the intelligent contract module is used for providing decentralized control function and key management function in the process of federal learning; the storage module based on the IPFS protocol is used for providing a decentralized information storage mechanism for the intermediate information in the federal learning process; and simultaneously operating the model training module, the intelligent contract module based on the block chain technology and the storage module based on the IPFS protocol on each node participating in the federal learning. The complete decentralization of the whole system is realized, the failure and exit of any node cannot influence other nodes to continue to carry out federal learning, and the robustness is stronger.
Description
Technical Field
The invention relates to the technical field of federal learning, in particular to a block chain-based federal learning system and a block chain-based federal learning method.
Background
Federated learning enables individual participating institutions to collaboratively train machine learning models without directly exchanging raw data. For enterprises or organizations with insufficient data volume, the data volume can be combined, a better model can be obtained, original data are not exposed, and mutual benefits and win-win can be realized. In the existing engineering technology, the cooperative training of each organization depends on a centralized third-party cooperative node to realize control, aggregation and key management. The existing centralization method has the following defects:
1) the cooperative node will continuously obtain the information uploaded by all other organizations. And a curious cooperative node can deduce important information related to the original data of each organization through the information, such as category label distribution, and therefore the privacy of the data can be leaked. They do not want to expose these privacy to the institutions involved in the training.
2) When the cooperative node fails, the whole system crashes and cannot continue to operate. Due to the single point of failure of the centralized cooperative node, the federal learning of each organization is forcibly terminated, and the cooperative training cannot be continued.
The federal learning method in the prior art has the problems of privacy risks and single point of failure.
Disclosure of Invention
The invention provides a block chain-based federal learning system and a block chain-based federal learning method, which aim to solve the existing problems.
In order to solve the above problems, the technical solution adopted by the present invention is as follows:
a blockchain-based federated learning system, comprising: the model training module is used for updating a machine learning model in the federal learning process and aggregating the change values of the machine learning model; the intelligent contract module is used for providing decentralized control function and key management function in the process of federal learning; the storage module based on the IPFS protocol is used for providing a decentralized information storage mechanism for the intermediate information in the federal learning process; and simultaneously operating the model training module, the intelligent contract module based on the block chain technology and the storage module based on the IPFS protocol on each node participating in the federal learning.
Preferably, the intelligent contract module based on the blockchain technology comprises a training control module and a key management module; the training control module is used for randomly generating a topological structure containing each node participating in the federal learning before each round of learning in the federal learning process, communicating with each node in the federal learning process, notifying the existing aggregation information of each node, and collecting the information of each node after further aggregation; the key management module stores public key information uploaded by each node participating in federal learning.
Preferably, the public key is a public key for homomorphic encryption; and each public key corresponds to each node participating in the federal learning one by one and is recorded by a key management module before the start of the federal learning.
Preferably, each of the nodes undertakes an aggregation task, and the aggregation of all the nodes together is the aggregated information of the block chain-based federated learning system.
Preferably, each node obtains encryption information from a predecessor node, and adds local encryption information to the obtained encryption information to obtain new encryption information; the new encryption information is transmitted to a subsequent node; and decrypting the accumulated encrypted information by the last node and updating the global training model to obtain the latest global model.
The invention provides a block chain-based federal learning method, which comprises the following steps: s1: all nodes participating in federal learning register public keys in a key management module of an intelligent contract module based on a block chain technology, and negotiate the structure, initial parameter information and the maximum training round number of a training model with each other; s2: the node carries out local training on the training model by using local data and records the change value of the training model; the node acquires a homomorphic encrypted public key used by the training of the current round from the key management module and encrypts a variation value of the training model; s3: the node determines the position of the node in the current topological structure according to the topological structure information in the topological cache; s4: the nodes sequentially accumulate and aggregate the encrypted change values of the training models under the notification of a training control module; s5: after the aggregation of the nodes is finished, the last node on the topological structure decrypts the aggregation information and updates the global model parameter; s6: each node acquires the latest global model parameter and updates the training model; s7: and the nodes are trained circularly until the training model converges or the maximum number of training rounds negotiated in advance is reached, and the training is stopped at the moment.
Preferably, the number of training samples of the local training is decided by the node; the topological structure information in the topological cache is generated by the training control module before each round of training is started, and the topological structure information in the topological cache marks the sequence of aggregation of the nodes.
Preferably, the nodes accumulate the encrypted information in sequence along the sequence under the notification of the training control module; and each node takes on the task of aggregation, and the aggregation of all the nodes is the aggregation information of the block chain-based federated learning system.
Preferably, before each round of training is started, the training control module selects a random node, and uses the node as the last node of the topology structure randomly generated in the current round, so as to decrypt the aggregated information and update the global model; if the selected node fails before the training of the current round begins, the training control module reselects one node; and if the selected node fails during the updating of the global model of the current round, the training control module informs all the nodes to roll back the training model to the state before the training of the current round is started, and then randomly selects one node again.
Preferably, in each round of training, if the node fails, the training control module skips the failed node, and the nodes immediately succeeding the failed node continue aggregation.
The invention has the beneficial effects that: the block chain-based federal learning system and method are provided, and by means of realizing decentralized of each module in federal learning, complete decentralized of the whole system is further realized. The status of each participating node is therefore completely equal and there is no centralized third party cooperative node. The failure and exit of any node can not influence other nodes to continue the federal learning, and compared with the existing method, the method has stronger robustness.
A decentralized encryption aggregation method is designed, so that no node can obtain information such as gradients or models of other nodes, and the risk of data privacy disclosure is avoided.
Due to the public transparency of the blockchain, the control logic of the whole federal learning process is also public transparency; the control logic is realized by an intelligent contract based on a block chain, and any node can check the intelligent contract, so that the whole control logic is safe and reliable; at the same time, due to the decentralized nature of the blockchain, the execution of the control logic is also decentralized; a blockchain-based intelligent dating implementing control logic is performed on all nodes participating in federal learning, and any single node cannot affect its operation.
Due to the fact that the block chain is not tamper-proof, historical information recorded in the whole federal learning process is recorded in the block chain and cannot be modified, and therefore auditing of behaviors of all nodes in the federal learning process is facilitated.
Drawings
Fig. 1 is a schematic structural diagram of a block chain-based federal learning system in an embodiment of the present invention.
Fig. 2 is a model diagram of a block chain-based federal learning system in an embodiment of the present invention.
Fig. 3 is a model diagram of a prior art federal learning system in an embodiment of the present invention.
Fig. 4 is a schematic diagram of a block chain-based federal learning method in an embodiment of the present invention.
FIG. 5 is a diagram illustrating the training of the block chain based federated learning system and method of embodiments of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the embodiments of the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and the embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It will be understood that when an element is referred to as being "secured to" or "disposed on" another element, it can be directly on the other element or be indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element. In addition, the connection may be for either a fixing function or a circuit connection function.
It is to be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like are used in an orientation or positional relationship indicated in the drawings for convenience in describing the embodiments of the present invention and to simplify the description, and are not intended to indicate or imply that the referenced device or element must have a particular orientation, be constructed in a particular orientation, and be in any way limiting of the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present invention, "a plurality" means two or more unless specifically limited otherwise.
As shown in fig. 1, the present invention provides a block chain-based federal learning system, which includes:
the model training module is used for updating a machine learning model in the federal learning process and aggregating the change values of the machine learning model;
the intelligent contract module is used for providing decentralized control function and key management function in the process of federal learning;
the storage module based on the IPFS protocol is used for providing a decentralized information storage mechanism for the intermediate information in the federal learning process;
and simultaneously operating the model training module, the intelligent contract module based on the block chain technology and the storage module based on the IPFS protocol on each node participating in the federal learning.
As shown in fig. 2 and 3, the present invention provides a decentralized, privacy-preserving federated learning system for institutions participating in training that enables the participating institutions to not exchange local raw data, to not leak data privacy, and to not rely on centralized nodes of other third parties. Participating institutions train in coordination with each other to obtain machine learning models that are better than their individual training results.
Furthermore, the accuracy of the model obtained by federal learning is ensured, and meanwhile, the data privacy of participating institutions is protected. Compared with the prior most advanced federal learning method, the method can achieve the same high model accuracy. Meanwhile, a new training method is designed, and information transmission in the training process is improved, so that the data privacy of participating mechanisms is effectively protected, and the risk of data privacy disclosure in the existing method is avoided.
Furthermore, the robustness of the whole system is ensured. Because the training control module of the system is realized based on the block chain technology design, and the IPFS protocol on which the storage module of the system is based is also a decentralized and distributed protocol, the normal operation of the system is maintained by all organizations, and the other organizations are not influenced to continue to carry out federal learning due to the quitting or the failure of any organization. Each participating authority within the system is perfectly equal and they can all be considered nodes. The architecture of the whole system is completely decentralized and only comprises the corresponding nodes of all the participating institutions, and the centralized nodes of any third party are not provided. Compared with the prior art, the block chain-based federated learning system has stronger robustness.
In one embodiment of the invention, the intelligent contract module based on the block chain technology comprises a training control module and a key management module; the training control module is used for randomly generating a topological structure containing each node participating in the federal learning before each learning in the federal learning process, communicating with each node in the federal learning process, informing the existing aggregation information of each node, and collecting the information of each node after further aggregation; the key management module stores the public key information uploaded by each node participating in the federal learning. The public key is a public key for homomorphic encryption; and each public key corresponds to each node participating in the federal learning one by one and is recorded by a key management module before the start of the federal learning. Each node undertakes the task of aggregation, and the aggregation of all the nodes is the aggregation information of the block chain-based federated learning system. Each node acquires encryption information from a precursor node, and adds the local encryption information and the acquired encryption information to obtain new encryption information; the new encryption information is transmitted to a subsequent node; and the accumulated encrypted information is decrypted by the last node and is used for updating the global training model to obtain the latest global model.
Then, the global model information is distributed to all nodes participating in the training, and the next iteration is started. In the training, the information decrypted by the final node is the accumulation of all the node encrypted information, and the information of any single node cannot be extracted from the information. Therefore, the training method does not leak the gradient or model information of any node, and the risk of privacy leakage in the existing centralization method is avoided.
The global training model is updated by using the aggregation information, then each node can obtain the global latest training model, and the local training model of the node is updated by using the global training model, namely, the global training model parameters are assigned to the local training model parameters.
The block chain-based federated learning system inherits the decentralized characteristic of the block chain technology and the IPFS protocol, so that the normal operation of the system is maintained by all nodes together without depending on the centralized nodes, and the failure and exit of any single node cannot influence other nodes to continue the federated learning. Compared with the existing method, the system has extremely high robustness.
As shown in fig. 4, the present invention further provides a block chain-based federal learning method, which includes the following steps:
s1: all nodes participating in federal learning register public keys in a key management module of an intelligent contract module based on a block chain technology, and negotiate the structure, initial parameter information and the maximum training round number of a training model with each other;
in one embodiment of the invention, the number of training samples for local training is determined by the participating nodes.
S2: the node carries out local training on the training model by using local data and records the change value of the training model; the node acquires a homomorphic encrypted public key used by the training of the current round from the key management module and encrypts a variation value of the training model;
in an embodiment of the present invention, before each round of training begins, the training control module selects a random one of the nodes, and uses a homomorphic encryption public key corresponding to the node as a public key used in the round of encryption; the node is used as the last node of the topological structure randomly generated in the current round and used for decrypting the aggregated information and updating the global model; if the selected node fails before the training of the current round is started, the training control module reselects a node; and if the selected node fails during the updating of the global model of the current round, the training control module informs all the nodes to roll back the training model to the state before the training of the current round is started, and then randomly selects one node again. S3: the node determines the position of the node in the current topological structure according to the topological structure information in the topological cache;
in an embodiment of the present invention, the topology information in the topology cache is generated by the training control module of the intelligent contract module based on the blockchain technology before each round of training starts, and the topology information in the topology cache marks the order of aggregation of the model training modules of the nodes; the node corresponding to the homomorphic encryption public key used in the current round must be the last node of the topology structure in the current round.
S4: the nodes sequentially accumulate and aggregate the encrypted change values of the training models under the notification of a training control module;
in an embodiment of the present invention, the model training module sequentially accumulates the encrypted information along the sequence under the notification of the training control module; and the model training module of each node bears the task of aggregation, and the aggregation of all the nodes is the aggregation information of the block chain-based federated learning system.
S5: after the aggregation of the nodes is finished, the last node on the topological structure decrypts the aggregation information and updates the global model parameter;
in an embodiment of the present invention, the homomorphic encryption public key used for the encryption of the aggregated information is a public key uploaded by the last node in the topology of the current round, so that only the node can decrypt the aggregated information by using a local private key.
S6: each node acquires the latest global model parameter and updates the training model;
in an embodiment of the present invention, the global model information is stored in a storage module based on an IPFS protocol, and a corresponding index is uploaded to the training control module; the training control module informs all nodes participating in the training of the index; and the node acquires the latest model parameters from the storage module based on the IPFS protocol according to the index and updates the local model parameters.
S7: and the nodes are trained circularly until the training model converges or the maximum number of training rounds negotiated in advance is reached, and the training is stopped at the moment.
For many small businesses and organizations, the user data they have is often very small, and it is difficult to train good models individually. Some existing data protection regulations limit the behavior of directly exchanging data among mechanisms; at the same time, organizations also do not want to divulge information related to local data for reasons of business interest and privacy, etc.
As shown in fig. 5, each organization first iterates the local model using the local data, and after a certain number of samples are iterated, obtains the public key that is homomorphically encrypted in this round from the key management module, and encrypts the change value of the model. Then, each mechanism accumulates the encrypted information sequentially along the position in the topology, informed by the training control module. And finally, the mechanism selected by the training control module decrypts the accumulated information and updates the global model. The new global model is distributed to all institutions and the next iteration continues. The information in the training process is saved by the system, and the information cannot be changed once being saved because the system inherits the non-tamper property of the block chain technology. This can be used to audit the activities of various agencies during the training process, so there is no concern that agencies may make malicious activities during the training process. Through our system, each organization can cooperate together, and the model with better effect is trained together under the condition of not depending on third party collaborators and not revealing data privacy. The mutual benefits and win-win are realized while the data privacy is protected by each mechanism.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several equivalent substitutions or obvious modifications can be made without departing from the spirit of the invention, and all the properties or uses are considered to be within the scope of the invention.
Claims (10)
1. A block chain based federated learning system, comprising:
the model training module is used for updating a machine learning model in the federal learning process and aggregating the change values of the machine learning model;
the intelligent contract module is used for providing decentralized control function and key management function in the process of federal learning;
the storage module based on the IPFS protocol is used for providing a decentralized information storage mechanism for the intermediate information in the federal learning process;
and simultaneously operating the model training module, the intelligent contract module based on the block chain technology and the storage module based on the IPFS protocol on each node participating in the federal learning.
2. The blockchain-based federated learning system of claim 1, wherein the blockchain-technology-based smart contract module includes a training control module and a key management module;
the training control module is used for randomly generating a topological structure containing each node participating in the federal learning before each round of learning in the federal learning process, communicating with each node in the federal learning process, notifying the existing aggregation information of each node, and collecting the information of each node after further aggregation;
the key management module stores public key information uploaded by each node participating in federal learning.
3. The blockchain-based federated learning system of claim 2, wherein the public key is a public key used for homomorphic encryption; and each public key corresponds to each node participating in the federal learning one by one and is recorded by a key management module before the start of the federal learning.
4. The block chain-based federated learning system of claim 1, wherein each of the nodes undertakes an aggregation task, the aggregation of all of the nodes together being aggregated information for the block chain-based federated learning system.
5. The block chain-based federated learning system of claim 4, wherein each of the nodes obtains encryption information from a predecessor node and adds local encryption information to the obtained encryption information to obtain new encryption information; the new encryption information is transmitted to a subsequent node; and decrypting the accumulated encrypted information by the last node and updating the global training model to obtain the latest global model.
6. A block chain-based federal learning method is characterized by comprising the following steps:
s1: all nodes participating in federal learning register public keys in a key management module of an intelligent contract module based on a block chain technology, and negotiate the structure, initial parameter information and the maximum training round number of a training model with each other;
s2: the node carries out local training on the training model by using local data and records the change value of the training model; the node acquires a homomorphic encrypted public key used by the training of the current round from the key management module and encrypts a variation value of the training model;
s3: the node determines the position of the node in the current topological structure according to the topological structure information in the topological cache;
s4: the nodes sequentially accumulate and aggregate the encrypted change values of the training models under the notification of the training control module;
s5: after the aggregation of the nodes is finished, the last node on the topological structure decrypts the aggregation information and updates the global model parameter;
s6: each node acquires the latest global model parameter and updates the training model;
s7: and the nodes are trained circularly until the training model converges or the maximum number of training rounds negotiated in advance is reached, and the training is stopped at the moment.
7. The blockchain-based federated learning method of claim 6, wherein the number of training samples of the local training is determined by the node;
the topological structure information in the topological cache is generated by the training control module before each round of training is started, and the topological structure information in the topological cache marks the sequence of aggregation of the nodes.
8. The block chain-based federated learning method of claim 7, wherein the nodes accumulate encryption information sequentially along a precedence order under the notification of the training control module; and each node takes on the task of aggregation, and the aggregation of all the nodes is the aggregation information of the block chain-based federated learning system.
9. The block chain-based federated learning method of claim 8, wherein before each round of training begins, the training control module selects a random one of the nodes as a last node of a topology randomly generated for the current round for decrypting the aggregated information and updating a global model; if the selected node fails before the training of the current round begins, the training control module reselects one node; and if the selected node fails during the updating of the global model of the current round, the training control module informs all the nodes to roll back the training model to the state before the training of the current round is started, and then randomly selects one node again.
10. The block chain-based federated learning method of claim 8, wherein during each round of training, if the node fails, the training control module skips the failed node and continues aggregation by the immediate successor of the failed node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911285920.1A CN111212110B (en) | 2019-12-13 | 2019-12-13 | Block chain-based federal learning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911285920.1A CN111212110B (en) | 2019-12-13 | 2019-12-13 | Block chain-based federal learning system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111212110A true CN111212110A (en) | 2020-05-29 |
| CN111212110B CN111212110B (en) | 2022-06-03 |
Family
ID=70787048
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911285920.1A Active CN111212110B (en) | 2019-12-13 | 2019-12-13 | Block chain-based federal learning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111212110B (en) |
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN111552986A (en) * | 2020-07-10 | 2020-08-18 | 鹏城实验室 | Block chain-based federal modeling method, device, equipment and storage medium |
| CN111652382A (en) * | 2020-06-04 | 2020-09-11 | 深圳前海微众银行股份有限公司 | Data processing method, device and equipment based on block chain 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 |
| CN111814985A (en) * | 2020-06-30 | 2020-10-23 | 平安科技（深圳）有限公司 | Model training method under federated learning network and related equipment thereof |
| CN111914038A (en) * | 2020-09-21 | 2020-11-10 | 北京百度网讯科技有限公司 | Federal calculation method, device, equipment and storage 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 |
| CN112183764A (en) * | 2020-10-12 | 2021-01-05 | 中国石油大学(华东) | Internet of things equipment fault detection method based on block chain and federal learning |
| CN112182102A (en) * | 2020-09-23 | 2021-01-05 | 西安纸贵互联网科技有限公司 | Method and device for processing data in federal learning, electronic equipment and storage medium |
| CN112232527A (en) * | 2020-09-21 | 2021-01-15 | 北京邮电大学 | Safe distributed federal deep learning method |
| CN112307331A (en) * | 2020-10-14 | 2021-02-02 | 湖南天河国云科技有限公司 | Block chain-based college graduate intelligent recruitment information pushing method and system and terminal equipment |
| CN112487103A (en) * | 2020-12-25 | 2021-03-12 | 昆明理工大学 | Trusted deep learning data set sharing system based on intelligent contract of block chain |
| CN112560088A (en) * | 2020-12-11 | 2021-03-26 | 同盾控股有限公司 | Knowledge federation-based data security exchange method and device and storage medium |
| 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 |
| CN112671746A (en) * | 2020-12-17 | 2021-04-16 | 杭州趣链科技有限公司 | Block chain-based federated learning model poisoning detection method |
| CN112733967A (en) * | 2021-03-30 | 2021-04-30 | 腾讯科技（深圳）有限公司 | Model training method, device, equipment and storage medium for federal learning |
| 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 |
| CN112769766A (en) * | 2020-12-23 | 2021-05-07 | 广东电网有限责任公司梅州供电局 | Safe aggregation method and system for data of power edge internet of things based on federal learning |
| 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 |
| CN112860800A (en) * | 2021-02-22 | 2021-05-28 | 深圳市星网储区块链有限公司 | Trusted network application method and device based on block chain and federal learning |
| CN112863175A (en) * | 2020-12-31 | 2021-05-28 | 平安科技（深圳）有限公司 | Automobile road monitoring data processing method, device, equipment and storage medium |
| CN113011598A (en) * | 2021-03-17 | 2021-06-22 | 深圳技术大学 | Financial data information federal migration learning method and device based on block chain |
| CN113052331A (en) * | 2021-02-19 | 2021-06-29 | 北京航空航天大学 | Block chain-based Internet of things personalized federal learning method |
| CN113111124A (en) * | 2021-03-24 | 2021-07-13 | 广州大学 | Block chain-based federal learning data auditing system and method |
| CN113159279A (en) * | 2021-03-18 | 2021-07-23 | 中国地质大学（北京） | Cross-domain knowledge assistance method and system based on neural network and deep learning |
| 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 |
| CN113191503A (en) * | 2021-05-20 | 2021-07-30 | 清华大学深圳国际研究生院 | Decentralized distributed learning method and system for non-shared data |
| CN113283778A (en) * | 2021-06-09 | 2021-08-20 | 哈尔滨工程大学 | Layered convergence federated learning method based on security evaluation |
| CN113326947A (en) * | 2021-05-28 | 2021-08-31 | 山东师范大学 | Joint learning model training method and system |
| CN113344221A (en) * | 2021-05-10 | 2021-09-03 | 上海大学 | Federal learning method and system based on neural network architecture search |
| CN113364589A (en) * | 2021-08-10 | 2021-09-07 | 深圳致星科技有限公司 | Key management system, method and storage medium for federal learning security audit |
| CN113408746A (en) * | 2021-06-22 | 2021-09-17 | 深圳大学 | Block chain-based distributed federal learning method and device and terminal equipment |
| CN113469371A (en) * | 2021-07-01 | 2021-10-01 | 建信金融科技有限责任公司 | Federal learning method and device |
| CN113609508A (en) * | 2021-08-24 | 2021-11-05 | 上海点融信息科技有限责任公司 | Block chain-based federal learning method, device, equipment and storage medium |
| CN113645197A (en) * | 2021-07-20 | 2021-11-12 | 华中科技大学 | Decentralized federal learning method, device and system |
| CN114326697A (en) * | 2020-10-09 | 2022-04-12 | 丰田自动车株式会社 | Model learning system, vehicle control device, and model learning method |
| CN114338144A (en) * | 2021-12-27 | 2022-04-12 | 杭州趣链科技有限公司 | Method for preventing data from being leaked, electronic equipment and computer-readable storage medium |
| WO2022174533A1 (en) * | 2021-02-20 | 2022-08-25 | 平安科技（深圳）有限公司 | Federated learning method and apparatus based on self-organized cluster, device, and storage medium |
| CN115190028A (en) * | 2022-06-16 | 2022-10-14 | 华中科技大学 | Decentralized federal learning method, device and system based on local area communication network |
| WO2022222152A1 (en) * | 2021-04-23 | 2022-10-27 | Oppo广东移动通信有限公司 | Federated learning method, federated learning system, first device, and third device |
| WO2023120802A1 (en) * | 2021-12-21 | 2023-06-29 | 경희대학교 산학협력단 | Federated learning system and operation method thereof |
| US11755954B2 (en) | 2021-03-11 | 2023-09-12 | International Business Machines Corporation | Scheduled federated learning for enhanced search |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180018590A1 (en) * | 2016-07-18 | 2018-01-18 | NantOmics, Inc. | Distributed Machine Learning Systems, Apparatus, and Methods |
| US20190303790A1 (en) * | 2018-03-27 | 2019-10-03 | Oben, Inc. | Proof of work based on training of machine learning models for blockchain networks |
| US20190334716A1 (en) * | 2018-04-27 | 2019-10-31 | The University Of Akron | Blockchain-empowered crowdsourced computing system |
| CN110472745A (en) * | 2019-08-06 | 2019-11-19 | 深圳前海微众银行股份有限公司 | Information transferring method and device in a kind of federal study |
| 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 |
| US20190370634A1 (en) * | 2018-06-01 | 2019-12-05 | International Business Machines Corporation | Data platform to protect security of data used by machine learning models supported by blockchain |
-
2019
- 2019-12-13 CN CN201911285920.1A patent/CN111212110B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180018590A1 (en) * | 2016-07-18 | 2018-01-18 | NantOmics, Inc. | Distributed Machine Learning Systems, Apparatus, and Methods |
| US20190303790A1 (en) * | 2018-03-27 | 2019-10-03 | Oben, Inc. | Proof of work based on training of machine learning models for blockchain networks |
| US20190334716A1 (en) * | 2018-04-27 | 2019-10-31 | The University Of Akron | Blockchain-empowered crowdsourced computing system |
| US20190370634A1 (en) * | 2018-06-01 | 2019-12-05 | International Business Machines Corporation | Data platform to protect security of data used by machine learning models supported by blockchain |
| CN110472745A (en) * | 2019-08-06 | 2019-11-19 | 深圳前海微众银行股份有限公司 | Information transferring method and device in a kind of federal study |
| 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 (2)
| Title |
|---|
| 杨强: "AI与数据隐私保护：联邦学习的破解之道", 《信息安全研究》 * |
| 杨强: "AI与数据隐私保护：联邦学习的破解之道", 《信息安全研究》, 30 November 2019 (2019-11-30) * |
Cited By (66)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652382B (en) * | 2020-06-04 | 2021-08-06 | 深圳前海微众银行股份有限公司 | Data processing method, device and equipment based on block chain and storage medium |
| CN111652382A (en) * | 2020-06-04 | 2020-09-11 | 深圳前海微众银行股份有限公司 | Data processing method, device and equipment based on block chain 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 |
| 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 |
| CN111814985A (en) * | 2020-06-30 | 2020-10-23 | 平安科技（深圳）有限公司 | Model training method under federated learning network and related equipment thereof |
| CN111814985B (en) * | 2020-06-30 | 2023-08-29 | 平安科技（深圳）有限公司 | Model training method under federal learning network and related equipment thereof |
| CN111552986A (en) * | 2020-07-10 | 2020-08-18 | 鹏城实验室 | Block chain-based federal modeling method, device, equipment and storage 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 |
| 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 |
| CN111914038B (en) * | 2020-09-21 | 2024-04-16 | 北京百度网讯科技有限公司 | Federal computing method, apparatus, device, and storage medium |
| CN111914038A (en) * | 2020-09-21 | 2020-11-10 | 北京百度网讯科技有限公司 | Federal calculation method, device, equipment and storage medium |
| CN112182102A (en) * | 2020-09-23 | 2021-01-05 | 西安纸贵互联网科技有限公司 | Method and device for processing data in federal learning, electronic equipment and storage medium |
| CN114326697B (en) * | 2020-10-09 | 2023-12-26 | 丰田自动车株式会社 | Model learning system, vehicle control device, and model learning method |
| CN114326697A (en) * | 2020-10-09 | 2022-04-12 | 丰田自动车株式会社 | Model learning system, vehicle control device, and model learning method |
| CN112183764A (en) * | 2020-10-12 | 2021-01-05 | 中国石油大学(华东) | Internet of things equipment fault detection method based on block chain and federal learning |
| CN112307331B (en) * | 2020-10-14 | 2023-11-24 | 湖南天河国云科技有限公司 | Intelligent recruitment information pushing method, system and terminal equipment for college graduates based on blockchain |
| CN112307331A (en) * | 2020-10-14 | 2021-02-02 | 湖南天河国云科技有限公司 | Block chain-based college graduate intelligent recruitment information pushing method and system and terminal equipment |
| 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 |
| 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 |
| CN112560088B (en) * | 2020-12-11 | 2024-05-28 | 同盾控股有限公司 | Knowledge federation-based data security exchange method, device and storage medium |
| CN112560088A (en) * | 2020-12-11 | 2021-03-26 | 同盾控股有限公司 | Knowledge federation-based data security exchange method and device and storage medium |
| CN112671746A (en) * | 2020-12-17 | 2021-04-16 | 杭州趣链科技有限公司 | Block chain-based federated learning model poisoning detection method |
| CN112769766A (en) * | 2020-12-23 | 2021-05-07 | 广东电网有限责任公司梅州供电局 | Safe aggregation method and system for data of power edge internet of things based on federal learning |
| CN112769766B (en) * | 2020-12-23 | 2023-03-24 | 广东电网有限责任公司梅州供电局 | Safe aggregation method and system for data of power edge internet of things based on federal learning |
| CN112487103A (en) * | 2020-12-25 | 2021-03-12 | 昆明理工大学 | Trusted deep learning data set sharing system based on intelligent contract of block chain |
| CN112487103B (en) * | 2020-12-25 | 2023-06-06 | 昆明理工大学 | Deep learning data set trusted sharing system based on blockchain intelligent contract |
| 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 |
| CN112863175B (en) * | 2020-12-31 | 2022-11-22 | 平安科技（深圳）有限公司 | Automobile road monitoring data processing method, device, equipment and storage medium |
| CN112863175A (en) * | 2020-12-31 | 2021-05-28 | 平安科技（深圳）有限公司 | Automobile road monitoring data processing method, device, equipment and storage medium |
| CN113052331A (en) * | 2021-02-19 | 2021-06-29 | 北京航空航天大学 | Block chain-based Internet of things personalized federal learning method |
| WO2022174533A1 (en) * | 2021-02-20 | 2022-08-25 | 平安科技（深圳）有限公司 | Federated learning method and apparatus based on self-organized cluster, device, and storage medium |
| CN112860800A (en) * | 2021-02-22 | 2021-05-28 | 深圳市星网储区块链有限公司 | Trusted network application method and device based on block chain and federal learning |
| US11755954B2 (en) | 2021-03-11 | 2023-09-12 | International Business Machines Corporation | Scheduled federated learning for enhanced search |
| CN113011598A (en) * | 2021-03-17 | 2021-06-22 | 深圳技术大学 | Financial data information federal migration learning method and device based on block chain |
| CN113159279B (en) * | 2021-03-18 | 2023-06-23 | 中国地质大学（北京） | Cross-domain knowledge assistance method and system based on neural network and deep learning |
| CN113159279A (en) * | 2021-03-18 | 2021-07-23 | 中国地质大学（北京） | Cross-domain knowledge assistance method and system based on neural network and deep learning |
| CN113111124A (en) * | 2021-03-24 | 2021-07-13 | 广州大学 | Block chain-based federal learning data auditing system and method |
| CN112733967B (en) * | 2021-03-30 | 2021-06-29 | 腾讯科技（深圳）有限公司 | Model training method, device, equipment and storage medium for federal learning |
| CN112733967A (en) * | 2021-03-30 | 2021-04-30 | 腾讯科技（深圳）有限公司 | Model training method, device, equipment and storage medium for federal learning |
| 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 |
| CN112801307B (en) * | 2021-04-13 | 2021-07-06 | 深圳索信达数据技术有限公司 | Block chain-based federal learning method and device and computer equipment |
| 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 |
| 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 |
| WO2022222152A1 (en) * | 2021-04-23 | 2022-10-27 | Oppo广东移动通信有限公司 | Federated learning method, federated learning system, first device, and third device |
| CN113344221A (en) * | 2021-05-10 | 2021-09-03 | 上海大学 | Federal learning method and system based on neural network architecture search |
| CN113191503A (en) * | 2021-05-20 | 2021-07-30 | 清华大学深圳国际研究生院 | Decentralized distributed learning method and system for non-shared data |
| CN113191503B (en) * | 2021-05-20 | 2023-06-09 | 清华大学深圳国际研究生院 | Decentralized distributed learning method and system for non-shared data |
| CN113326947A (en) * | 2021-05-28 | 2021-08-31 | 山东师范大学 | Joint learning model training method and system |
| CN113326947B (en) * | 2021-05-28 | 2023-06-16 | 山东师范大学 | Training method and system for joint learning model |
| CN113283778A (en) * | 2021-06-09 | 2021-08-20 | 哈尔滨工程大学 | Layered convergence federated learning method based on security evaluation |
| CN113283778B (en) * | 2021-06-09 | 2023-08-01 | 哈尔滨工程大学 | Layered convergence federal learning method based on security evaluation |
| 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 |
| CN113469371A (en) * | 2021-07-01 | 2021-10-01 | 建信金融科技有限责任公司 | Federal learning method and device |
| CN113645197A (en) * | 2021-07-20 | 2021-11-12 | 华中科技大学 | Decentralized federal learning method, device and system |
| CN113645197B (en) * | 2021-07-20 | 2022-04-29 | 华中科技大学 | Decentralized federal learning method, device and system |
| CN113364589A (en) * | 2021-08-10 | 2021-09-07 | 深圳致星科技有限公司 | Key management system, method and storage medium for federal learning security audit |
| CN113609508B (en) * | 2021-08-24 | 2023-09-26 | 上海点融信息科技有限责任公司 | Federal learning method, device, equipment and storage medium based on blockchain |
| CN113609508A (en) * | 2021-08-24 | 2021-11-05 | 上海点融信息科技有限责任公司 | Block chain-based federal learning method, device, equipment and storage medium |
| WO2023120802A1 (en) * | 2021-12-21 | 2023-06-29 | 경희대학교 산학협력단 | Federated learning system and operation method thereof |
| CN114338144A (en) * | 2021-12-27 | 2022-04-12 | 杭州趣链科技有限公司 | Method for preventing data from being leaked, electronic equipment and computer-readable storage medium |
| CN115190028A (en) * | 2022-06-16 | 2022-10-14 | 华中科技大学 | Decentralized federal learning method, device and system based on local area communication network |
| CN115190028B (en) * | 2022-06-16 | 2024-05-14 | 华中科技大学 | Decentralized federal learning method, device and system based on local area communication network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111212110B (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111212110B (en) | Block chain-based federal learning system and method | |
| CN111209334B (en) | Power terminal data security management method based on block chain | |
| US11102185B2 (en) | Blockchain-based service data encryption methods and apparatuses | |
| CN108646983B (en) | Processing method and device for storing service data on block chain | |
| CN102625939B (en) | System and method for managing electronic assets | |
| CN102696045B (en) | System and method for performing serialization of devices | |
| Halder | A cyber security for a smart grid | |
| CN106982203A (en) | The ATM network system and its information processing method of robust based on block chain technology | |
| CN107203344A (en) | A kind of date storage method and data-storage system | |
| CN102648471A (en) | System and method for hardware based security | |
| Zhao et al. | Private and verifiable interdomain routing decisions | |
| CN109547488A (en) | A kind of trust data calculating and exchange system based on alliance's block chain | |
| CN114500111B (en) | Multi-platform-based automatic project audit data processing method and system | |
| CN110035058B (en) | Resource request method, device and storage medium | |
| CN113469716A (en) | Product tracing method and system based on block chain and identification analysis | |
| CN110866261A (en) | Data processing method and device based on block chain and storage medium | |
| CN109858272A (en) | Blacklist data processing system based on block chain | |
| CN110929229A (en) | Block chain-based office document credibility verification method and system | |
| CN110909381A (en) | Block chain-based user comment tamper-proof system and method | |
| CN113704353A (en) | Block chain credit investigation method integrating information chain and privacy chain | |
| CN113556393B (en) | Multi-type intermodal data exchange system and method based on block chain | |
| CN109873828A (en) | Method for managing resource, device, equipment and storage medium | |
| CN109728908A (en) | A kind of key management method based on quantum safety moving storage medium | |
| CN115953244A (en) | Transaction supervision method and device based on block chain, electronic equipment and storage medium | |
| CN109509007A (en) | Product false proof electronic tag and electronic tag authentication method and system |
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 |