Founding Entity: Qatar National Research Fund (QNRF)
Amount of the Founding: $ 1M+
Duration of the Project: 3 years
Lead Investigator: Prof. Dr. Roberto Di Pietro
Fintech applications are massively distributed and need to process a large amount of data. To support these applications, multiple cloud service providers and network service providers need to collaborate with banks, financial institutions, and other entities. This introduces several management and security challenges, since the network is open, entities do not trust each other, and compete with each other for same businesses. To resolve the problem, third-party trusted brokers are used to manage the resource exchanges between each pair of entities. For example, the Society for Worldwide Interbank Financial Telecommunication (SWIFT) is used for international transfers worldwide including Qatar. These third party brokers introduce extra cost and latency and become the single point of failure.
Blockchains provide a way for a global consensus among entities that do not trust each other. All parties can agree whether a transaction is valid or not valid. That is, the validity of a transaction is either 1 or 0. However, there are several situations, where a probabilistic consensus would be more appropriate. For example, whether investing in a particular stock will be profitable. Given a number of financial institutions, each will have its own answer. The overall consensus can be best expressed as a probability between 0 and 1. In this project we extend the blockchain technology to help reach such consensus.
Numerous other applications will benefit from the probabilistic blockchains. Financial institutions make risky investments. These investments need a probabilistic consensus, not a binary agreement. Insurance, currency exchange, commodity trading, oil and gas exploration are just a few examples where the investing partners would want to agree on the probability of success rather than on the certainty of it.
In this project we integrate several cutting-edge research fields, including blockchain, distributed systems, network security, cloud computing, and deep learning. First, we plan to design a framework for evaluating the suitability of currently available blockchain platforms in Fintech applications. We then plan to develop the probabilistic blockchain concept to achieve non-binary consensus in an efficient fashion. Then, we will integrate this with intelligent and adaptive IDS agents, built using deep learning techniques.
The proposed probabilistic blockchains will extend blockchains’ applicability beyond simple distributed ledger applications being investigated by the Fintech industry. It will be helpful in all risk-based investments that form the bulk of the Fintech business.
The outcome of this project will be several papers, patents, and reports on the probabilistic blockchains mechanisms for reaching consensus.