I had the pleasure to supervise a range of students for their master thesis, UROP, and group projects. A selection of projects can be found here. If you are interested on working on a project with me, please reach out.
Florian Apfelbeck’s master thesis project was to develop a privacy-preserving machine learning project. In this project, a Multi-Party Computation (MPC) protocol was used to realise a linear regression and a shallow neural network implementation without reveling input data of the participants in the protocol.
I worked with Manuel Zander and Tom Waite to create a simulator for decentralised ledgers that use a Directed Acyclic Graph (DAG) instead of the usual blockchain. We investigated their properties and our results are summarised in a paper published at SOCCA 2018.
Amine Kakouche created a neural network method to anlyse the ledger by the Hyperledger Indy project. The project showed that the ledger itself cannot be analysed using statistical methods, however, in an emulated environment learning on metadata of network packages is possible.
Six students at Imperial created a Token Curated Registry (TCR) simulator that allows designers to experiments with different parameters in TCRs. Their experiences are documented in a series of blog posts and give a good insight how software engineering projects at Imperial are executed.
Decentralised exchanges (DEX) promise to be safe against centralised providers having custody over funds and are an active area of research. A group of students build a version of a DEX using OrbitDB, a database build on IPFS using CRDTs to achieve strong eventual consistency. Exciting technologies!
Decentralized ledgers, such as Bitcoin and Ethereum, have gained rapid popularity, attracting the attention of academics, entrepreneurs, economists, and policy-makers. They promise and already create new disruptive markets, and revolutionize how we think of money and financial infrastructure. This course teaches the basics of distributed ledgers.