The sheer quantity and complexity of interactions in our society (human, governmental, political, financial) and critical infrastructures (power grids, transport, water & gas networks, telecom) is dramatically increasing. Algorithms and intelligent agents are therefore being delegated the management and control of these systems.

The ability to accurately discover hidden relations between items that share similarities in particular is of paramount importance to these agents. Clustering algorithms in particular have become prevalent: once clusters of similar items have been identified, subsequent analysis and optimization procedures benefit from a reduction in dimensionality.

In the past, clustering algorithms ignored time-dependent structures within data. Recently however, we made state-of-the-art advances into the detection of clusters in so-called Block Markov Chains (BMCs). One can now observe just one trajectory of a Markov chain, and provably recover hidden clusters, all using our algorithm.

In light of this exciting development, you will now:
(1) Develop time-dependent clustering methods suitable for application in real-world networks and datastreams in collaboration with KPN.
(2) Push the scope of application of time-dependent clustering procedures into new theoretical directions, leveraging inspiration gained from the collaboration.
(3) Merge reinforcement learning with clustering techniques on BMCs, and perform regret analyses to quantify the efficacy of your new learning algorithms.

To achieve (1)-(3) the Networks Architectures and Services (NAS) research group invites applicants for this challenging PhD position. Your position will be part of NExTWORKx, an exciting collaboration between academia and industry:

https://www.tudelft.nl/2018/tu-delft/kpn-en-tu-delft-gaan-samenwerken-aan-nieuwe-ict-technologieen

The ideal candidate is brilliant, has a strong interest in theoretically developing, understanding, and implementing learning / clustering algorithms on networks and datastreams, and is well-versed in probability theory, statistics, machine learning and network science.

Fixed-term contract: 4 years.

The TU Delft offers a customisable compensation package, a discount for health insurance and sport memberships, and a monthly work costs contribution. Flexible work schedules can be arranged. An International Children’s Centre offers childcare and an international primary school. Dual Career Services offers support to accompanying partners. Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities.

As a PhD candidate you will be enrolled in the TU Delft Graduate School. The TU Delft Graduate School provides an inspiring research environment; an excellent team of supervisors, academic staff and a mentor; and a Doctoral Education Programme aimed at developing your transferable, discipline-related and research skills. Please visit graduateschool.tudelft.nl/ for more information.

Faculty Electrical Engineering, Mathematics and Computer Science

The Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS) is known worldwide for its high academic quality and the social relevance of its research programmes. The faculty’s excellent facilities accentuate its international position in teaching and research. Within this interdisciplinary and international setting the faculty employs more than 1100 employees, including about 400 graduate students and about 2100 students. Together they work on a broad range of technical innovations in the fields of sustainable energy, telecommunications, microelectronics, embedded systems, computer and software engineering, interactive multimedia and applied mathematics.

The Networks Architectures and Services (NAS) research group conducts research in the broad area of complex networks, ranging from data communications and Internetworking to man-made infrastructures such as road-traffic, biological, brain, social, financial networks, and now, networks of quantum computational devices. Our strengths lie in understanding these complex networks through topological and spectral studies, as well as the stochastic modelling thereof.

https://www.tudelft.nl/en/eemcs/the-faculty/