According to research published this month in the Journal project.
The study introduces the Machine Learning Blockchain (MLOB), an innovative approach that addresses significant vulnerabilities in the current system that integrates artificial intelligence with blockchain networks.
Although machine learning has been rapidly adopted in the field of engineering to improve accuracy and efficiency, current blockchain solutions focus primarily on data security, making the actual computing process vulnerable to manipulation, a gap that researchers aim to shut down.
“The research goal is to develop a novel ML on the blockchain framework to ensure data and computing processes security,” authors Zhiming Dong and Weisheng Lu wrote in the paper. “The central purpose is to put them all on the blockchain, execute them as blockchain smart contracts, and protect execution records on the chain.”
Unlike existing systems where existing systems typically run machine learning models in off-chain environments, they are still susceptible to tampering, the MLOB framework embeds data and computational processes on the blockchain itself. This creates a safe, traceable environment for sensitive engineering calculations.
Safety without performance fines
The researchers tested their framework for multiple attack schemes and compared it with conventional methods. Their tests show that the MLOB framework successfully defended against all six attack scenarios designed to harm the system, surpassing all baseline methods.
Perhaps most surprisingly, enhanced security does not bring huge performance costs. Compared with the traditional method, the team’s evaluation only showed negligible differences in accuracy, while the delay of each calculation task increased by only 0.231 seconds.
“The key finding is that MLOB can significantly improve the computational security of engineering computing without increasing the computing power requirements,” the researchers noted. “This finding could alleviate concerns about the computing resource requirements for ML-BT integration.”
For industries where computing security is critical, such as critical infrastructure, quality assurance systems, or engineering forensics, this edge efficiency trade-off may be worth a significant improvement.
Real-world applications
To demonstrate practical applications, the team implemented the framework in the context of the construction industry and monitored the progress of interior buildings by comparing its construction conditions with planned models.
“The construction progress monitoring process and its results are highly correlated with progress payments and quality accountability,” the researchers explained. Noting that traditional monitoring is “vulnerable to potential threats or interferences, which may compromise the accuracy of the progress estimation process.”
In this real-world test, the framework provides verifiable security throughout the calculation process, maintaining high accuracy.
The system works by first acquiring and training machine learning models for specific tasks. The model is then converted into a format compatible with blockchain deployment and loaded it firmly into the blockchain, and ultimately ensures computational integrity through a consensus-based process.
Balancing safety, accuracy and efficiency
Researchers frame through what is called the “balance triangle” of security, accuracy and efficiency, and identify improvements in one area usually require trade-offs in other areas.
“Although this framework enhances computational security, it may incur costs in terms of efficiency and accuracy due to the occupied computing resources,” they wrote. However, their assessments suggest that “this rebalancing maintains a satisfactory level of efficiency and accuracy.”
The framework has reached a critical moment as the engineering field increasingly relies on artificial intelligence to accomplish complex tasks while facing growing security threats.
The impact of the industry is not just technological improvements. The researchers believe that the framework can drive innovation in engineering practices by integrating advanced technologies, which may lead to “more competitive engineering operations, increased productivity and attractiveness to talent interested in cutting-edge technologies.”
Future development
The system is not without limitations. Current implementations provide limited support for latency-sensitive applications and lack user-friendly interfaces, which are areas for future development.
“In order to improve the efficacy and user-friendliness of MLOB, future efforts should be used to expand the platform,” the researchers wrote.
For organizations focusing on computing security in engineering applications, researchers recommend starting with pilot exploration, comprehensive training programs, ongoing performance monitoring, and iterative improvements based on real-world feedback.
As machine learning continues to quickly integrate it into critical engineering systems, it may become increasingly important not only to ensure data, but also to ensure the computing process itself. The MLOB framework is an important step in solving these emerging security challenges.
The full text of Open Paper “Machine Learning on Blockchain (MLOB): A New Paradigm for Computational Security in Engineering” is provided in Journal Engineering.
If you find this piece useful, consider supporting our work with a one-time or monthly donation. Your contribution allows us to continue to bring you accurate, thought-provoking scientific and medical news that you can trust. Independent reporting requires time, effort, and resources, and your support makes it possible for us to continue exploring stories that are important to you. Together, we can ensure that important discoveries and developments attract those who need them the most.
