Blockchain Smart Contract Verification for Machine Learning Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for machine learning model verification and transaction lack secure computation methods to ensure confidentiality of data and models, and are vulnerable to fraud during transactions.
Innovation Solution
A method utilizing fully homomorphic encryption (FHE) and smart contracts on a blockchain system to securely verify machine learning models and facilitate transactions, ensuring confidentiality and preventing fraud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional verification methods are used, then verification can be performed, but data confidentiality and model security are compromised
Solution Approach 1:
The patent introduces a blockchain system as an intermediary to facilitate secure verification and transaction between model provider and buyer. The blockchain executes smart contracts that automatically verify model performance using encrypted data, eliminating the need for direct data sharing and preventing fraud. The intermediary blockchain coordinates the verification process, manages cryptocurrency transactions, and ensures both parties fulfill their obligations without compromising data security.
Solution Approach 2:
The patent replaces traditional mechanical verification processes with cryptographic homomorphic encryption and blockchain technology. Instead of manually verifying models with direct data access, the system uses homomorphic encryption to perform computations on encrypted data, allowing verification without decrypting sensitive information. This substitution of mechanical verification with cryptographic mechanisms ensures data confidentiality while maintaining verification reliability.
2Object-affected harmful factors
If secure computation methods are implemented, then data confidentiality is ensured, but computational complexity increases
Solution Approach 1:
The patent segments the verification process into distinct phases: data encryption, model encryption, verification execution, and result decryption. Each phase operates on encrypted data independently, allowing complex verification computations to be divided into manageable steps. The segmentation enables the system to handle cryptographic operations systematically, reducing the perceived complexity while maintaining security.
Solution Approach 2:
The blockchain system serves multiple functions simultaneously: it executes smart contracts, performs cryptographic verification, manages cryptocurrency transactions, and ensures data confidentiality. By making the verification system multi-functional, the patent reduces overall system complexity compared to having separate specialized systems for each function. The blockchain acts as a universal platform that handles verification, transaction, and security requirements.
3Ease of operation
If manual verification processes are used, then flexibility is maintained, but human resources are consumed and fraud is possible
Solution Approach 1:
The patent implements self-service verification where the blockchain system automatically executes verification contracts without human intervention. The smart contracts autonomously perform cryptographic operations, evaluate model performance, and manage transactions based on predefined conditions. This self-service mechanism maintains operational flexibility through programmable verification logic while eliminating the need for manual human verification processes.
Solution Approach 2:
The system incorporates feedback mechanisms where verification results automatically trigger corresponding transactions. When the blockchain verifies that a model meets performance criteria, it automatically releases cryptocurrency payments to the model provider. This feedback loop creates an automated verification-transaction system that eliminates manual intervention while maintaining flexibility through configurable verification parameters and smart contract logic.
4Object-affected harmful factors
If encrypted data is used for verification, then model security is improved, but decryption and model access become more difficult
Solution Approach 1:
The patent applies preliminary action by encrypting both the test data and model before verification begins. The blockchain system performs all verification computations using these pre-encrypted data, eliminating the need for decryption during the verification process. This preliminary encryption ensures model security throughout the entire verification pipeline while avoiding the complexity of decryption operations.
Solution Approach 2:
The blockchain system acts as an intermediary that handles encrypted data operations without requiring direct access to decrypted models. The smart contracts manage the encryption keys and coordination between parties, allowing secure model access only after verification succeeds. The intermediary blockchain layer abstracts the complexity of key management and decryption, providing simplified model access to authorized parties.
Data Source
AI summary
The present disclosure provides a method including: generating a first private key and a public key according to a parameter set of full homomorphic encryption; encrypting test data and label by the public key to generate test data ciphertext and label ciphertext; generating a smart contract executed by a blockchain system, and transferring control of an amount of cryptocurrency from a first cryptocurrency account to the blockchain; receiving a result of a verification to a model ciphertext; when the result indicates that the model ciphertext does not pass the verification, retrieving the control of the amount of cryptocurrency; and when the result indicates that the model ciphertext passes the verification, receiving the model ciphertext and a second private key from the blockchain system, and decrypting, according to the first and second private keys, the model ciphertext to generate a model to infer the test data.