CN111680793B - Block chain consensus method and system based on deep learning model training - Google Patents
Block chain consensus method and system based on deep learning model training Download PDFInfo
- Publication number
- CN111680793B CN111680793B CN202010318933.0A CN202010318933A CN111680793B CN 111680793 B CN111680793 B CN 111680793B CN 202010318933 A CN202010318933 A CN 202010318933A CN 111680793 B CN111680793 B CN 111680793B
- Authority
- CN
- China
- Prior art keywords
- file
- information
- uri
- deep learning
- node
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/382—Payment protocols; Details thereof insuring higher security of transaction
- G06Q20/3825—Use of electronic signatures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a block chain consensus method and system based on deep learning model training, wherein the system comprises a parameter and data acquirer, a consensus algorithm dispatcher, a deep learning model trainer and a consensus verifier; can be used for interfacing with the existing blockchain system to replace the consensus algorithms such as POW/POS/POA. According to the invention, excessive computing power of the blockchain is introduced into the training of the deep learning model, and through an excitation mechanism of the blockchain, an investor can use the mining machine for training the artificial intelligent model, and can guide investment of funds, computing power and energy into more meaningful work, so that the problems of insufficient computing power and high cost are solved. The POW computing power of the block chain is used for carrying out the training calculation of the deep learning model of big data, so that the cost is reduced, the social resource is saved, and the computing power is used for meaningful work.
Description
Technical Field
The invention relates to the technical field of blockchains, in particular to a blockchain consensus method and system based on deep learning model training.
Background
The development of the blockchain technology is widely accepted by domestic and foreign enterprises, research institutions, universities and the like, and is considered to be the core of the next-generation value Internet. The main stream blockchain platform bit coin, the Ethernet, the Lai coin and the like generally adopt a POW consensus algorithm, and the nodes of the block are determined by repeatedly carrying out hash value operation through the mining machine, so that the investment of which node has higher calculation power and has higher probability to obtain the block rewards. The mechanism enables trust establishment and value consensus to be achieved through machines and algorithms, investors invest a large amount of money to purchase ore machines and electric power, by 2019, 9 months, the computing power of the whole bitcoin net is up to 70EH/S, the computing power and energy are used by investors to carry out hash value operation to obtain rewards, and the waste of computing power and energy in the mode is greatly affected.
Deep Learning (DL) is a new research direction in the field of Machine Learning (ML), which was introduced to Machine Learning to bring it closer to the original goal-artificial intelligence (AI, artificial Intelligence). Deep learning is the inherent regularity and presentation hierarchy of learning sample data, and the information obtained during such learning is helpful in interpreting data such as text, images and sounds. Its final goal is to have the machine have analytical learning capabilities like a person, and to recognize text, image, and sound data. Deep learning is a complex machine learning algorithm that achieves far greater results in terms of speech and image recognition than prior art. According to one statistic, the computational effort used in artificial intelligence training tasks has been exponentially increasing since 2012, at current rates of doubling every 3.5 months (in contrast, moore's law doubling every 18 months). Since 2012, the demand for computing power has increased by more than 300,000 times (and only 12 times if at the rate of moore's law). During this time, the improvement of hardware computing power has been an important factor in the rapid development of artificial intelligence. Thus, computational effort plays a vital role in the training of deep learning models.
The mainstream blockchain generally adopts a common-knowledge algorithm such as POW and the like to determine block nodes, so that the problems of energy and calculation power waste exist.
Disclosure of Invention
In view of the above, in order to solve the problem of energy and computational power waste in the block chain consensus algorithm in the prior art, the invention provides a block chain consensus method and system based on deep learning model training.
The invention solves the problems by the following technical means:
in one aspect, the invention provides a blockchain consensus method based on deep learning model training, comprising the following steps:
the main body collects the file information and returns the uniform resource locator URI after storing the file information in the file system File The principal constructs a data collection transaction, collects the reward amount, URI, according to the format { principal wallet address ] File Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node;
the subject obtains the storedThe file information is marked, a marked file is generated and stored in a file system, and then a uniform resource locator URI is returned Labeling The principal constructs a data tagging transaction, tags the bonus amount, URI, according to a format { principal wallet address ] File ,URI Labeling Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node; wherein URI is Labeling The corresponding file is a pair URI File The corresponding file is marked with information;
training a consensus algorithm by adopting a deep learning model to calculate and verify, and verifying the validity of the transaction by using an accounting node, and putting the transaction into a buffer pool until the number n of data sets A in the buffer pool reaches a threshold value alpha;
the billing node acquires all data annotation transactions stored in a chain and all data annotation transactions in a node buffer pool by using a parameter and data acquirer, acquires files and annotation information corresponding to a URI file and URI annotation, acquires a parameter value stored in a previous block or a previous block, initializes a deep learning neural network by using the parameter value, starts supervised learning, and automatically adjusts a network structure and parameters by adopting an AutoML method until the prediction accuracy of a model is greater than a threshold value beta set by a system;
the accounting node completes the calculation of the model, stores model parameters into a block header, generates a first block cast transaction for recording that the node obtains accounting rewards, packages the transactions in a buffer pool together into a block body, combines the block header and the block body to generate a block and performs full network broadcasting, and searches for verification signatures of other verification nodes;
the other verification nodes receive the information of the new block, and the 'consensus verifier' verifies the information; when the deep learning consensus model is adopted, a verification node acquires all data labeling transactions stored in a chain by using a parameter and data acquirer, acquires a file T and labeling information F corresponding to a URI file and a URI label, acquires parameter values stored in the block, predicts T after initializing a deep learning neural network by using the parameter values, compares with F, and calculates a correctness threshold gamma of the model; if the threshold gamma requirement is met, the verification node signs the threshold gamma requirement and returns to the accounting node;
after the accounting node receives the signature of the verification node with a certain threshold proportion delta, the accounting node sends the block information again and attaches the signatures of all the verification nodes;
and after receiving the information of the new block and verifying that the signature information in the new block passes, the other nodes add the requirement into the block chain.
Further, before the deep learning model training consensus algorithm is adopted for calculation and verification, the method further comprises the following steps:
the "consensus algorithm scheduler" of the billing node selects the consensus algorithm, if the consensus algorithm is trained for the deep learning model, the consensus algorithm is trained using the deep learning model, otherwise, the calculation and verification is performed using the conventional blockchain consensus algorithm.
Further, the file information includes pictures, voice, video, and text.
Further, the underlying blockchain if based on bitcoin, lat, blockchain would extend the new data collection type, here the "URI File "reserved field in the bit coin transaction is used for storage; if the block chain platform with intelligent contract mechanism based on Ethernet, EOS, URI File The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is the acquired rewarding amount.
Further, the underlying blockchain if based on bitcoin, lat, blockchain would extend the new data collection type, here the "URI File +URI Labeling "reserved field in the bit coin transaction is used for storage; if the block chain platform with intelligent contract mechanism based on Ethernet, EOS, URI File +URI Labeling The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes the intelligent contract verification, the data information passes the intelligent contract verification and then passes the intelligent contract to go to the main walletThe address "transfer, the amount is" mark rewards amount ".
Further, the thresholds α, β, γ, δ are determined by intelligent contractual means and voting by all node participants.
In another aspect, the present invention provides a blockchain consensus system based on deep learning model training, comprising:
parameter and data acquirer for main body collecting file information and returning uniform resource locator URI after storing it in file system File The principal constructs a data collection transaction, collects the reward amount, URI, according to the format { principal wallet address ] File Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node; the main body obtains the stored file information, marks the file information, generates a marked file, stores the marked file in a file system and returns a uniform resource locator URI Labeling The principal constructs a data tagging transaction, tags the bonus amount, URI, according to a format { principal wallet address ] File ,URI Labeling Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node; wherein URI is Labeling The corresponding file is a pair URI File The corresponding file is marked with information;
the deep learning model trainer is used for calculating and verifying by adopting a deep learning model training consensus algorithm, and the accounting node verifies the validity of the transaction and puts the transaction into the buffer pool until the number n of the data sets A in the buffer pool reaches a threshold value alpha; the billing node acquires all data annotation transactions stored in a chain and all data annotation transactions in a node buffer pool by using a parameter and data acquirer, acquires files and annotation information corresponding to a URI file and URI annotation, acquires a parameter value stored in a previous block or a previous block, initializes a deep learning neural network by using the parameter value, starts supervised learning, and automatically adjusts a network structure and parameters by adopting an AutoML method until the prediction accuracy of a model is greater than a threshold value beta set by a system; the accounting node completes the calculation of the model, stores model parameters into a block header, generates a first block cast transaction for recording that the node obtains accounting rewards, packages the transactions in a buffer pool together into a block body, combines the block header and the block body to generate a block and performs full network broadcasting, and searches for verification signatures of other verification nodes;
the common identification verifier is used for verifying the information of the new block received by other verification nodes; when the deep learning consensus model is adopted, a verification node acquires all data labeling transactions stored in a chain by using a parameter and data acquirer, acquires a file T and labeling information F corresponding to a URI file and a URI label, acquires parameter values stored in the block, predicts T after initializing a deep learning neural network by using the parameter values, compares with F, and calculates a correctness threshold gamma of the model; if the threshold gamma requirement is met, the verification node signs the threshold gamma requirement and returns to the accounting node; after the accounting node receives the signature of the verification node with a certain threshold proportion delta, the accounting node sends the block information again and attaches the signatures of all the verification nodes; and after receiving the information of the new block and verifying that the signature information in the new block passes, the other nodes add the requirement into the block chain.
Further, the blockchain consensus system trained based on the deep learning model further comprises:
and the consensus algorithm scheduler is used for selecting a consensus algorithm by the accounting node, if the consensus algorithm is trained for the deep learning model, the deep learning model is used for training the consensus algorithm to calculate and verify, otherwise, the traditional block chain consensus algorithm is used for calculating and verifying.
Further, the underlying blockchain if based on bitcoin, lat, blockchain would extend the new data collection type, here the "URI File "reserved field in the bit coin transaction is used for storage; if the block chain platform with intelligent contract mechanism based on Ethernet, EOS, URI File The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the money is acquired rewarding money;
underlying blockchains such asIf the result is based on the blockchain of bitcoin and Lai te coin, then a new data collection type will be extended, here a URI File +URI Labeling "reserved field in the bit coin transaction is used for storage; if the block chain platform with intelligent contract mechanism based on Ethernet, EOS, URI File +URI Labeling The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is marked with rewarding amount.
Further, the file information comprises pictures, voice, video and text; the thresholds α, β, γ, δ are determined by intelligent contractual means and voting by all node participants.
Compared with the prior art, the invention has the beneficial effects that at least:
aiming at big data such as pictures, videos, voices and texts, the invention provides a method for training a deep learning model by using the data as a block chain consensus, and can apply calculation force to valuable model mining.
According to the invention, excessive computing power of the blockchain is introduced into the training of the deep learning model, and through an excitation mechanism of the blockchain, an investor can use the mining machine for training the artificial intelligent model, and can guide investment of funds, computing power and energy into more meaningful work, so that the problems of insufficient computing power and high cost are solved.
The invention uses the POW computing power of the block chain to carry out the training calculation of the deep learning model of big data, reduces the cost, saves the social resource and uses the computing power to meaningful work.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a blockchain consensus method of the present invention based on deep learning model training;
FIG. 2 is a schematic diagram of a block chain consensus system based on deep learning model training in accordance with the present invention.
Detailed Description
In order to make the above objects, features and advantages of the present invention more comprehensible, the following detailed description of the technical solution of the present invention refers to the accompanying drawings and specific embodiments. It should be noted that the described embodiments are only some embodiments of the present invention, and not all embodiments, and that all other embodiments obtained by persons skilled in the art without making creative efforts based on the embodiments in the present invention are within the protection scope of the present invention.
Example 1
As shown in FIG. 1, the invention provides a block chain consensus method based on deep learning model training, comprising the following steps:
step S110: the main body collects the file information such as pictures, voice, video, text and the like, stores the file information in a file system and returns a uniform resource locator URI File The principal constructs a data collection transaction, collects the reward amount, URI, according to the format { principal wallet address ] File And the node broadcasts the information to neighboring nodes.
The underlying blockchain if based on the blockchain of bitcoin, lai te coin, etc., would extend the new data collection type, here the URI File "reserved field in token transaction can be used for storage; if the blockchain platform with intelligent contract mechanism is based on Ethernet, EOS and the like, URI File The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is the acquired rewarding amount.
Step S120: the main body obtains the stored file information such as pictures, voice, video, text and the like, marks the file information, and generates marksReturning uniform resource locator URI after notes file is stored in file system Labeling The principal constructs a data tagging transaction, tags the bonus amount, URI, according to a format { principal wallet address ] File ,URI Labeling And the node broadcasts the information to neighboring nodes. Wherein URI is Labeling The corresponding file is a pair URI File Corresponding files such as pictures, voice, video, text and the like are marked.
The underlying blockchain if based on the blockchain of bitcoin, lai te coin, etc., would extend the new data collection type, here the URI File +URI Labeling "reserved field in token transaction can be used for storage; if the blockchain platform with intelligent contract mechanism is based on Ethernet, EOS and the like, URI File +URI Labeling The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is marked with rewarding amount.
The data sources may not come from data information stored on the chain, but rather, the training and validation of the model may be performed by specifying a training set of data and a testing set by way of task publishing.
Step S130: the "consensus algorithm scheduler" of the billing node selects the consensus algorithm, if the consensus algorithm is trained for the deep learning model, then jumps to step S140, otherwise the calculation and verification is performed using the conventional blockchain consensus algorithm.
Step S140: the accounting node verifies the validity of the transaction and puts it in the buffer pool until the number n of data sets a in the buffer pool reaches the threshold a.
Step S150: the billing node acquires all data labeling transactions stored in a chain and all data labeling transactions in a node buffer pool by using a parameter and data acquirer, acquires files and labeling information corresponding to a URI file and a URI label, acquires a parameter value stored in a previous block or a previous block, initializes a deep learning neural network by using the parameter value, starts supervised learning, and automatically adjusts a network structure and parameters by adopting a AutoML (Automated Machine Learning) method until the prediction accuracy of a model is greater than a threshold value beta set by a system.
Step S160: the accounting node completes the calculation of the model, stores the model parameters into the block header, generates a first block cast transaction for recording that the node obtains accounting rewards, packages the transactions in the buffer pool together into a block body, combines the block header and the block body to generate a block and performs full network broadcasting, and searches for verification signatures of other verification nodes.
Step S170: the other verification nodes receive the information of the new block, and the common identification verifier verifies the information. When the deep learning consensus model is adopted, the verification node acquires all data labeling transactions stored in a chain by using a parameter and data acquirer, acquires a file T and labeling information F corresponding to a URI file and a URI label, acquires parameter values stored in the block, predicts T after initializing a deep learning neural network by using the parameter values, compares the parameter values with F, and calculates a series of thresholds gamma such as accuracy of the model. If the threshold gamma requirement is reached, the authentication node signs it and returns to the accounting node.
Step S180, when the accounting node receives the signature of the verification node with a certain threshold proportion delta, the accounting node sends the block information again and attaches the signatures of all the verification nodes.
Step S190, after receiving the information of the new block and verifying that the signature information therein passes, the other nodes add the requirement to the block chain.
The thresholds α, β, γ, δ, etc. may be determined by means of smart contracts, etc. and voted by all node participants.
Example 2
As shown in FIG. 2, the invention further provides a block chain consensus system based on deep learning model training, which comprises a parameter and data acquirer, a consensus algorithm scheduler, a deep learning model trainer and a consensus verifier.
Parameter and data acquirer for main body to acquire picture, voice and visionAfter storing file information such as frequency, text and the like in a file system, returning a uniform resource locator URI File The principal constructs a data collection transaction, collects the reward amount, URI, according to the format { principal wallet address ] File And the node broadcasts the information to neighboring nodes.
The underlying blockchain if based on the blockchain of bitcoin, lai te coin, etc., would extend the new data collection type, here the URI File "reserved field in token transaction can be used for storage; if the blockchain platform with intelligent contract mechanism is based on Ethernet, EOS and the like, URI File The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is the acquired rewarding amount.
The main body obtains the stored file information such as pictures, voice, video, text and the like, marks the file information, generates a marked file, stores the marked file in a file system and returns a uniform resource locator URI Labeling The principal constructs a data tagging transaction, tags the bonus amount, URI, according to a format { principal wallet address ] File ,URI Labeling And the node broadcasts the information to neighboring nodes. Wherein URI is Labeling The corresponding file is a pair URI File Corresponding files such as pictures, voice, video, text and the like are marked.
The underlying blockchain if based on the blockchain of bitcoin, lai te coin, etc., would extend the new data collection type, here the URI File +URI Labeling "reserved field in token transaction can be used for storage; if the blockchain platform with intelligent contract mechanism is based on Ethernet, EOS and the like, URI File +URI Labeling The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is marked with rewarding amount.
The data sources may not come from data information stored on the chain, but rather, the training and validation of the model may be performed by specifying a training set of data and a testing set by way of task publishing.
And the consensus algorithm scheduler is used for selecting a consensus algorithm by the accounting node, if the consensus algorithm is trained for the deep learning model, the deep learning model is used for training the consensus algorithm to calculate and verify, otherwise, the traditional block chain consensus algorithm is used for calculating and verifying.
The deep learning model trainer is used for verifying the validity of the transaction by the accounting node, and putting the transaction into the buffer pool until the number n of the data sets A in the buffer pool reaches a threshold value alpha.
The billing node acquires all data labeling transactions stored in a chain and all data labeling transactions in a node buffer pool by using a parameter and data acquirer, acquires files and labeling information corresponding to a URI file and a URI label, acquires a parameter value stored in a previous block or a previous block, initializes a deep learning neural network by using the parameter value, starts supervised learning, and automatically adjusts a network structure and parameters by adopting a AutoML (Automated Machine Learning) method until the prediction accuracy of a model is greater than a threshold value beta set by a system.
The accounting node completes the calculation of the model, stores the model parameters into the block header, generates a first block cast transaction for recording that the node obtains accounting rewards, packages the transactions in the buffer pool together into a block body, combines the block header and the block body to generate a block and performs full network broadcasting, and searches for verification signatures of other verification nodes.
And the consensus verifier is used for verifying the information of the new block received by other verification nodes. When the deep learning consensus model is adopted, the verification node acquires all data labeling transactions stored in a chain by using a parameter and data acquirer, acquires a file T and labeling information F corresponding to a URI file and a URI label, acquires parameter values stored in the block, predicts T after initializing a deep learning neural network by using the parameter values, compares the parameter values with F, and calculates a series of thresholds gamma such as accuracy of the model. If the threshold gamma requirement is reached, the authentication node signs it and returns to the accounting node.
When the accounting node receives the signature of the verification node for a threshold proportion delta, the accounting node again transmits the block information and attaches the signatures of all verification nodes.
And after receiving the information of the new block and verifying that the signature information in the new block passes, the other nodes add the requirement into the block chain.
The thresholds α, β, γ, δ, etc. may be determined by means of smart contracts, etc. and voted by all node participants.
The invention provides a method and a system for achieving block chain node consensus by training a deep learning model, which are divided into a consensus algorithm dispatcher, a parameter and data acquirer, a deep learning model trainer and a consensus verifier. The system can be in butt joint with the existing block chain system to replace the consensus algorithm such as POW/POS/POA and the like.
Aiming at big data such as pictures, videos, voices and texts, the invention provides a method for training a deep learning model by using the data as a block chain consensus, and can apply calculation force to valuable model mining.
According to the invention, excessive computing power of the blockchain is introduced into the training of the deep learning model, and through an excitation mechanism of the blockchain, an investor can use the mining machine for training the artificial intelligent model, and can guide investment of funds, computing power and energy into more meaningful work, so that the problems of insufficient computing power and high cost are solved.
The invention uses the POW computing power of the block chain to carry out the training calculation of the deep learning model of big data, reduces the cost, saves the social resource and uses the computing power to meaningful work.
The foregoing examples illustrate only a few embodiments of the invention and are described in detail herein without thereby limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (10)
1. A block chain consensus method based on deep learning model training is characterized by comprising the following steps:
the main body collects the file information and returns the uniform resource locator URI after storing the file information in the file system File The principal constructs a data collection transaction, collects the reward amount, URI, according to the format { principal wallet address ] File Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node;
the main body obtains the stored file information, marks the file information, generates a marked file, stores the marked file in a file system and returns a uniform resource locator URI Labeling The principal constructs a data tagging transaction, tags the bonus amount, URI, according to a format { principal wallet address ] File ,URI Labeling Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node; wherein URI is Labeling The corresponding file is a pair URI File The corresponding file is marked with information;
training a consensus algorithm by adopting a deep learning model to calculate and verify, and verifying the validity of the transaction by using an accounting node, and putting the transaction into a buffer pool until the number n of data sets A in the buffer pool reaches a threshold value alpha;
the accounting node acquires all data labeling transactions stored in a chain and all data labeling transactions in a node buffer pool by using a parameter and data acquirer, acquires files and labeling information corresponding to a URI file and a URI label, acquires a parameter value stored in a previous block or a previous block, initializes a deep learning neural network by using the parameter value, starts supervised learning, and automatically adjusts a network structure and parameters by using an AutoML method until the prediction accuracy of a model is greater than a threshold value beta set by a system;
the accounting node completes the calculation of the model, stores model parameters into a block header, generates a first block cast transaction for recording that the node obtains accounting rewards, packages the transactions in a buffer pool together into a block body, combines the block header and the block body to generate a block and performs full network broadcasting, and searches for verification signatures of other verification nodes;
other verification nodes receive the information of the new block, and the consensus verifier verifies the information; when the deep learning consensus model is adopted, the verification node acquires all data labeling transactions stored in a chain by using parameters and a data acquirer, acquires a file T and labeling information F corresponding to a URI file and URI labeling, acquires parameter values stored in the block, predicts T after initializing a deep learning neural network by using the parameter values, compares with F, and calculates a correctness threshold gamma of the model; if the threshold gamma requirement is met, the verification node signs the threshold gamma requirement and returns to the accounting node;
after the accounting node receives the signature of the verification node with a certain threshold proportion delta, the accounting node sends the block information again and attaches the signatures of all the verification nodes;
and after receiving the information of the new block and verifying that the signature information in the new block passes, the other nodes add the block into the block chain.
2. The deep learning model training based blockchain consensus method according to claim 1, further comprising, prior to calculating and verifying using the deep learning model training consensus algorithm:
and the consensus algorithm dispatcher of the accounting node selects a consensus algorithm, if the consensus algorithm is trained for the deep learning model, the consensus algorithm is trained by adopting the deep learning model, otherwise, the calculation and verification are carried out by adopting the traditional block chain consensus algorithm.
3. The deep learning model training based blockchain consensus method of claim 1, wherein the file information includes pictures, voice, video and text.
4. The deep learning model training-based blockchain consensus method as in claim 1Wherein the underlying blockchain, if based on bitcoin, lai-coin blockchain, will extend the new data collection type, here URI File Storing by adopting reserved fields in the bitcoin transaction; URI if the Ethernet, EOS based blockchain platform with Smart contract mechanism File The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to the main wallet address through the intelligent contract, and the amount is the collected rewarding amount.
5. The deep learning model training based blockchain consensus method as in claim 1, wherein the underlying blockchain, if based on bitcoin, laite blockchain, will extend a new data collection type, here a "URI File +URI Labeling "reserved field in the bit coin transaction is used for storage; if the block chain platform with intelligent contract mechanism based on Ethernet, EOS, URI File +URI Labeling The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to the main wallet address through the intelligent contract, and the amount is the marked rewarding amount.
6. The deep learning model training based blockchain consensus method according to claim 1, wherein the thresholds α, β, γ, δ are determined by intelligent contractual means and votes by all node participants.
7. A blockchain consensus system based on deep learning model training, comprising:
parameter and data acquirer for main body collecting file information and returning uniform resource locator URI after storing it in file system File The principal constructs a data collection transaction, collects the reward amount, URI, according to the format { principal wallet address ] File Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node; subject access to storageThe file information of (2) is marked, a marked file is generated and is stored in a file system, and then a uniform resource locator (URI) is returned Labeling The principal constructs a data tagging transaction, tags the bonus amount, URI, according to a format { principal wallet address ] File ,URI Labeling Transmitting the information to the node, and broadcasting the information to adjacent nodes by the node; wherein URI is Labeling The corresponding file is a pair URI File The corresponding file is marked with information;
the deep learning model trainer is used for calculating and verifying by adopting a deep learning model training consensus algorithm, and the accounting node verifies the validity of the transaction and puts the transaction into the buffer pool until the number n of the data sets A in the buffer pool reaches a threshold value alpha; the accounting node acquires all data labeling transactions stored in a chain and all data labeling transactions in a node buffer pool by using a parameter and data acquirer, acquires files and labeling information corresponding to a URI file and a URI label, acquires a parameter value stored in a previous block or a previous block, initializes a deep learning neural network by using the parameter value, starts supervised learning, and automatically adjusts a network structure and parameters by using an AutoML method until the prediction accuracy of a model is greater than a threshold value beta set by a system; the accounting node completes the calculation of the model, stores model parameters into a block header, generates a first block cast transaction for recording that the node obtains accounting rewards, packages the transactions in a buffer pool together into a block body, combines the block header and the block body to generate a block and performs full network broadcasting, and searches for verification signatures of other verification nodes;
the common identification verifier is used for verifying the information of the new block received by other verification nodes; when the deep learning consensus model is adopted, the verification node acquires all data labeling transactions stored in a chain by using parameters and a data acquirer, acquires a file T and labeling information F corresponding to a URI file and URI labeling, acquires parameter values stored in the block, predicts T after initializing a deep learning neural network by using the parameter values, compares with F, and calculates a correctness threshold gamma of the model; if the threshold gamma requirement is met, the verification node signs the threshold gamma requirement and returns to the accounting node; after the accounting node receives the signature of the verification node with a certain threshold proportion delta, the accounting node sends the block information again and attaches the signatures of all the verification nodes; and after receiving the information of the new block and verifying that the signature information in the new block passes, the other nodes add the block into the block chain.
8. The deep learning model training based blockchain consensus system of claim 7, further comprising:
and the consensus algorithm scheduler is used for selecting a consensus algorithm by the accounting node, if the consensus algorithm is trained for the deep learning model, the deep learning model is used for training the consensus algorithm to calculate and verify, otherwise, the traditional block chain consensus algorithm is used for calculating and verifying.
9. The deep learning model training based blockchain consensus system as in claim 7, wherein the underlying blockchain would extend the new data collection type if the blockchain were bit coin, lat coin based, where URI File Storing by adopting reserved fields in the bitcoin transaction; URI if the Ethernet, EOS based blockchain platform with Smart contract mechanism File The method is called by taking the data information as a parameter and calling the intelligent contract, and when the data information passes through the intelligent contract verification, the data information is transferred to a main wallet address through the intelligent contract, and the amount is the collected rewarding amount;
the underlying blockchain if based on bitcoin, lai-coin blockchain will extend the new data collection type, here the URI File +URI Labeling "reserved field in the bit coin transaction is used for storage; if the block chain platform with intelligent contract mechanism based on Ethernet, EOS, URI File +URI Labeling The method is called by taking the data information as a parameter and calling the intelligent contract, and after the data information passes the intelligent contract verification, the data information passes the intelligent contract to the mainAnd transferring the money box address, wherein the money amount is marked with the rewarding amount.
10. The deep learning model training based blockchain consensus system of claim 7, wherein the file information includes pictures, voice, video, and text; the thresholds α, β, γ, δ are determined by intelligent contractual means and voting by all node participants.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010318933.0A CN111680793B (en) | 2020-04-21 | 2020-04-21 | Block chain consensus method and system based on deep learning model training |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010318933.0A CN111680793B (en) | 2020-04-21 | 2020-04-21 | Block chain consensus method and system based on deep learning model training |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111680793A CN111680793A (en) | 2020-09-18 |
CN111680793B true CN111680793B (en) | 2023-06-09 |
Family
ID=72451667
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010318933.0A Active CN111680793B (en) | 2020-04-21 | 2020-04-21 | Block chain consensus method and system based on deep learning model training |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111680793B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114282586A (en) * | 2020-09-27 | 2022-04-05 | 中兴通讯股份有限公司 | Data annotation method, system and electronic equipment |
CN112364388A (en) * | 2020-10-28 | 2021-02-12 | 中车工业研究院有限公司 | Sensor data authentication method and device based on block chain |
CN112418335B (en) * | 2020-11-27 | 2024-04-05 | 北京云聚智慧科技有限公司 | Model training method based on continuous image frame tracking annotation and electronic equipment |
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 |
CN112819072B (en) * | 2021-02-01 | 2023-07-18 | 西南民族大学 | Supervision type classification method and system |
CN112967148B (en) * | 2021-03-30 | 2024-02-20 | 电子科技大学长三角研究院(衢州) | Block chain consensus mechanism for intelligent Internet of things computing service |
CN113254980B (en) * | 2021-07-07 | 2022-02-15 | 南京可信区块链与算法经济研究院有限公司 | Workload certification consensus method and system for deep learning |
CN113837300B (en) * | 2021-09-29 | 2024-03-12 | 上海海事大学 | Automatic driving cross-domain target detection method based on block chain |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107864198A (en) * | 2017-11-07 | 2018-03-30 | 济南浪潮高新科技投资发展有限公司 | A kind of block chain common recognition method based on deep learning training mission |
CN109543726A (en) * | 2018-11-06 | 2019-03-29 | 联动优势科技有限公司 | A kind of method and device of training pattern |
CN110197285A (en) * | 2019-05-07 | 2019-09-03 | 清华大学 | Security cooperation deep learning method and device based on block chain |
CN110535836A (en) * | 2019-08-12 | 2019-12-03 | 安徽师范大学 | A kind of trust block chain common recognition method of based role classification |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190012595A1 (en) * | 2017-07-07 | 2019-01-10 | Pointr Data, Inc. | Neural network consensus using blockchain |
-
2020
- 2020-04-21 CN CN202010318933.0A patent/CN111680793B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107864198A (en) * | 2017-11-07 | 2018-03-30 | 济南浪潮高新科技投资发展有限公司 | A kind of block chain common recognition method based on deep learning training mission |
CN109543726A (en) * | 2018-11-06 | 2019-03-29 | 联动优势科技有限公司 | A kind of method and device of training pattern |
CN110197285A (en) * | 2019-05-07 | 2019-09-03 | 清华大学 | Security cooperation deep learning method and device based on block chain |
CN110535836A (en) * | 2019-08-12 | 2019-12-03 | 安徽师范大学 | A kind of trust block chain common recognition method of based role classification |
Non-Patent Citations (2)
Title |
---|
Meng Shen等.Privacy-Preserving Support Vector Machine Training Over Blockchain-Based Encrypted IoT Data in Smart Cities.IEEE Internet of Things Journal.2019,第7702-7712页. * |
方俊杰 ; 雷凯 ; .面向边缘人工智能计算的区块链技术综述.应用科学学报.2020,(第01期),第1-21页. * |
Also Published As
Publication number | Publication date |
---|---|
CN111680793A (en) | 2020-09-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111680793B (en) | Block chain consensus method and system based on deep learning model training | |
CN112862298B (en) | Credit evaluation method for user portrait | |
CN111681091A (en) | Financial risk prediction method and device based on time domain information and storage medium | |
CN112541817A (en) | Marketing response processing method and system for potential customers of personal consumption loan | |
CN111680098A (en) | Block chain system for data acquisition, data annotation, AI model training and verification | |
CN115293910A (en) | Intelligent enterprise cash flow rating system based on financial big data | |
CN113011646A (en) | Data processing method and device and readable storage medium | |
CN111695335A (en) | Intelligent interviewing method and device and terminal equipment | |
CN110046345A (en) | A kind of data extraction method and device | |
CN114626102A (en) | Block chain-based electronic certificate transfer method, device, equipment and storage medium | |
CN111061948A (en) | User label recommendation method and device, computer equipment and storage medium | |
US20240095457A1 (en) | Systems and methods for generating dynamic conversational responses based on predicted user intents using artificial intelligence models | |
Zeng et al. | Mitigating inconsistencies in multimodal sentiment analysis under uncertain missing modalities | |
CN112365352B (en) | Anti-cash-out method and device based on graph neural network | |
CN116861258B (en) | Model processing method, device, equipment and storage medium | |
CN113159796A (en) | Trade contract verification method and device | |
CN113362852A (en) | User attribute identification method and device | |
CN111680076A (en) | Block chain consensus method and system based on association rule model training | |
CN110414845B (en) | Risk assessment method and device for target transaction | |
Ali et al. | An intelligent model for success prediction of initial coin offerings | |
CN111680099B (en) | Block chain consensus method and system based on decision tree model training | |
CN116109318B (en) | Interactive financial payment and big data compression storage method and system based on blockchain | |
CN111178535B (en) | Method and apparatus for implementing automatic machine learning | |
CN117217807B (en) | Bad asset estimation method based on multi-mode high-dimensional characteristics | |
CN115809875A (en) | Cross-bank transfer risk identification method and device |
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 |