Fri 3 Jul 20
In recent years, blockchain technology – which allows digital information to be distributed between networks without being copied – has become increasingly more popular in the world of business.
A research paper by Somdip Dey, from our School of Computer Science and Electronic Engineering, was the first to propose a framework to secure against a majority attack in blockchain using machine learning and algorithmic game theory.
Many researchers around the world are now working on the topic of securing blockchain using machine learning, based on Somdip’s published work Securing Majority-Attack in Blockchain Using Machine Learning and Algorithmic Game Theory: A Proof of Work
We asked Somdip to explain a bit more about his research.
Blockchain is a revolutionary technology as it helps reduce risk and increase anonymity by offering a transparent, secure and fraud-resistant platform for a digital document to be distributed, which creates trust. Blockchain is popularly used in the form of crypto currency, however, recently the technology is also being used for many different forms of digital assets.
The blockchain market is now established and is expected to grow significantly and aggressively over the next few years. The security element is a key reason for that growth and effective strategies must be applied to that end. One key security concern in blockchain is the majority attack where several participants in a blockchain network can collude to gain control of the network, which is, in fact, a valid concern for consortium-based blockchain networks.
Recently we have seen several institutions coming together to create consortium-based blockchain networks such as Hyperledger. Although for applications of blockchain such as Bitcoin and Litcoin the majority-attack might not be a great threat, for consortium-based blockchain networks - where several institutions such as public, private, government, etc. are collaborating - the majority-attack might just prove to be a prevalent threat if collusion among these institutions takes place.
In my published work, we propose a methodology where we can use intelligent software agents to monitor the activity of stakeholders in the blockchain networks to detect an anomaly such as collusion, using supervised machine learning algorithm and algorithmic game theory and stop the majority-attack from taking place.
Recently, we have noticed an increase in start-ups and private institutions participating in developing their own blockchain solutions for digital assets and, therefore, the threat of a majority attack becomes more prevalent in such a scenario. Many researchers around the world are now working on the topic of securing blockchain using machine learning based on my published work.
Yes, we are trying to use blockchain in food security and securely implementing a blockchain network is a priority to gain trust among our users. Recently, I co-founded a food tech company, Nosh Technologies, which is a tech spin-out of the University of Essex, and we are trying to utilize blockchain in the food industry to track the supply chain of food products and make food accessible to individuals at a cheaper cost. Given the application of blockchain in the food industry, security plays an important factor in successfully implementing such a technology. My published research on this matter has been critical for our company’s development of this implementation of blockchain.
We have researchers at the School of Computer Science and Electronics Engineering (CSEE), especially researchers from the Centre for Computational Finance and Economic Agents (CCFEA), who are working on integrating blockchain, economics and finance.
Professor Klaus McDonald-Maier and I are exploring the security aspect of blockchain networks, where Professor McDonald-Maier’s research focuses on hardware/software security integration of blockchain and my research focuses on the algorithmic side of blockchain security.
Given the current state of the research in blockchain we could see self-healing blockchain networks in the near future, where the network is capable of identifying threats on its own without being supervised (without the use of supervised machine learning) and also capable of securing itself from such threats. If we reach such a level of security in blockchain then we could notice more institutes, especially government agencies, adopting blockchain for their target application. In that case I only hope that our research, pursued at Essex, plays an important role in the development of such a technology.