-
Notifications
You must be signed in to change notification settings - Fork 454
MasterThesis
You need to share our love for running code and operational systems.
All thesis projects are exploratory research projects 45 ECTS master thesis. We offer a very wide range of topics: Web3, decentralised learning, blockchains, digital identity, digital Euros, personal data vaults, digital trust, etc.
PageRank is the defining centralised algorithm of Google. A small field is investigating architectures for decentralised machine learning. One of the initial works in this field is from Delft, G-Rank: Unsupervised Continuous Learn-to-Rank for Edge Devices in a P2P Network. The goal of this master thesis is to understand, improve, and implement the first true decentralised machine learning algorithm.
As a general intro, read the Wikipedia entry on decentralised search engines. Early work from 2005 by Delft provides a simple and realistic experimental approach to the problem of media discovery and recommendation, you are required to understand the basic algorithm of semantic clustering (e.g. taste buddies). A paper from 2012 proposes a model where mobile phones use gossip learning to compute a linear model, without revealing local models and without storing the full data set. Another classic attempt from 2003 onwards is the decentralised YaCy search engine with a web crawler, complex hashing, and reverse word index.
See example: https://arxiv.org/search/?query=pouwelse+survey&searchtype=all&source=header Approach: read 30+ papers, put them in a table, call this a taxonomy, write code, make screenshot, include this, DONE.