MasterThesis
The Tribler lab takes inspiration from Linux, Wikipedia, Bittorrent, and Bitcoin. These technologies disrupted existing industries and brought change to the world. You thesis will contribute to something similar. All thesis projects are exploratory research projects of 45 ECTS. We offer a very wide range of topics: Web3, decentralised learning, blockchains, digital trust, digital identity, digital Euros, personal data vaults, etc.
See all ongoing and past master thesis projects on Github. Every master thesis is documented on Github, code is on Github, and all thesis drafts are on Github. Every progress meeting is documented in a few lines, we publish our findings immediately, providing open notebook science. Various open topics:
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. The application of your algorithm is Internet search or something else you prefer.
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.
The superapp by Delft is the starting point of a Web3 alternative for Big Tech.
You master thesis focus is on the key part of the Web3 technology stack: content storage and dissemination. Pay Bittorrent seeders in Bitcoin for their efforts. Create a self-organising and censorship-free content platform based on free-market principles.
We believe that the DAO is the future of the firm, providing economic activity without boundaries. See prior work by Delft: our unstoppable DAO. Your task is to further realise this dream of a robot economy or even the fourth industrial revolution. Key goals are collective money and democratic decision making processes.
You master thesis task is to design and deploy a Euro coin with true gold value. For several years we have building prototypes at Delft for the upcoming digital Euro. See the latest offline Euro work by a team of master students for 5 ECTS. By using a small amount of antique gold it is possible to back a digital stable coin. The GoldEuro is an ownership token for actual physical gold. As such, it is stable, well understood legally, and easy to deploy. Going back to the gold standard has significant implication for payment systems and central banks. Money again has intrinsic value, limiting commercial money creation by banks. With GoldEuro citizens take back power in the global economy.
Official TUDelft 10 ECTS course code and description
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. Template for the TUDelft literature Survey is as follows:
- Based on arXiv IEEE 2-clumn format in latex
- Indicate that this is not a classical scientific article, but student project
- First section Introduction
- Problem Description
- One or few section with the focus of the survey
- Literature survey has always 1 essential taxonomy table. Taxonomy in a single glance. Systematic overview of past years of innovation. Key milestones identified. Scientific grounding, 1 or more scientific article per entry/line/milestone. Table with overview and literature, see brilliant With Honours example:
Official TUDelft Requirements and timing in preparation for the Thesis assessment
- First steps
- Get "green light" from the Thesis Advisor
- Set the date for the defence
- 40 working days before the planned assessment (defence)
- Hand in final version of the thesis report (to Thesis Advisor). -30 working days before the planned assessment
- Submit the examination application via the ‘Final Examination Form’, signed by the responsible professor. The Board of Examiners checks fulfilment of all requirements. For SET you will find the Final Examination Form here. Make a reservation for a room/lecture hall.
- (At least) 20 working days before the planned assessment
- Send the final report to the members of the Thesis Committee and invite the Thesis Committee for the Assessment. Supply a digital or printed copy of the final Thesis report to the members of the Thesis Committee.
- 10 working days before the planned assessment
- The Thesis Committee sends her remarks to you and the thesis advisor.
- You send a clear announcement of the Thesis defence to the Thesis Committee members, to the staff of the department of the chosen specialisation with the title of the thesis, the place of the defence, the date and time, the committee, and the abstract of his project (250 words). If you wish, you may also invite family and friends.
- (At least) 5 working days before the planned assessment
- You incorporate the remarks into the Thesis report.
- You upload a pdf of the final version of the Thesis report in the electronic TU Delft repository.