PhD student for characterization procedures of near-term quantum computers

Quantum computing devices are rapidly increasing in scale and complexity. However they are still very far away from the abstract ideal of a computer, and an important part of making them better is designing procedures that characterize errors and noise in an efficient and robust. Do you enjoy working at the intersection between physics and computer science? Are you comfortable with mathematical work and interested in applying those skills in a practical setting? Then CWI invites you to apply for a PhD position where you will research characterization procedures for quantum devices, in the context of the Dutch Quantum Delta Initiative.

What are going to do?
As part of your PhD project, you will develop and mathematically analyze new procedures to characterize quantum computing devices and collaborate with experimental teams to apply them in real systems. Possible research directions include:

• Develop new characterization procedures based on the shadow estimation and randomized benchmarking techniques and help experimental teams with implementing them in practice.
• Develop analytical measures of non-Markovianity in noise processes and develop protocols to estimate them in practice
• Investigate the surprising noise tolerance of certain classes of quantum variational ansatzes (such as the DMERA tensor network ansatz)

You will work under the supervision of Dr. Jonas Helsen and be a part of QuSoft, the Dutch research institute for quantum software, which houses experts on all topics in quantum computing. Together, the local institutes form a rich research ecosystem with opportunities to work together with computer scientists, physicists and mathematicians from the various groups situated at the Amsterdam Science Park.

You will also closely collaborate with experimental teams at the QuTech institute at the TU Delft, giving you the opportunity to participate in the development of quantum computers up close.

You will:

• Research near-term quantum computation, by developing new characterization procedures and by mathematically analyzing their properties;
• Assist experimental teams in the application of these characterization procedures;
• Publish and present your work in international conferences and journals;
• Be an active part of the local research groups at CWI and QuSoft;

Phd student on Dynamic X-ray Computed Tomography using Deep Generative Networks

X-ray Computed Tomography (CT) is the most versatile 3D imaging technique to reveal the interior of static objects with high spatial resolution. It is indispensable in an increasing range of applications in science, health care and industry. However, despite considerable research, state-of-the-art (SOTA) image reconstruction methods fail to resolve dynamic processes with both high spatial and temporal resolution. This PhD position is part of an NWO Vidi project that will use deep generative neural networks to develop image reconstruction techniques that allow to realize the full potential of CT to reveal, analyse, control or optimize dynamical processes. Using a particular dynamic extension of deep generative image models for dynamic CT (called “dgdCT” hereafter) may circumvent most disadvantages of SOTA methods, but challenging methodical developments combined with experimental data acquisition and extensive evaluation are needed to prove this. The project will develop dgdCT, publish the first open software implementation that scales to high-resolution real-world applications, and provide the imaging community with an unprecedented, open, experimental data collection used for evaluation and benchmarking. Demonstrations on important real-world applications will showcase the potential of dgdCT to boost dynamic CT imaging. The project relies on the unique combination of the Computational Imaging group’s expertise in translating sophisticated image reconstruction techniques to real-world applications, a network of collaborations and our group’s software developments, experimental facilities and computational resources.

Within the Vidi project, the PhD student will work together with the PI Felix Lucka on methods development: First, we will develop a software implementation of dgdCT that is efficient enough to scale to challenging high-resolution applications while still being flexible enough to allow for curiosity-driven methodical research. Then, we will examine how to accelerate the numerical optimization behind fitting a dgdCT to new data so that it becomes a viable method for challenging large-scale applications. Finally, we will develop extensions of dgdCT that improve it and unlock its potential for tasks beyond image reconstruction. The methods developed will be tested on our group’s computational clusters for high-performance GPU computing and evaluated with simulated and experimental X-ray CT data.

The PhD candidate will be supervised by Felix Lucka, co-supervised by Tristan van Leeuwen and will be part of a large, dynamic research group together with other PhD candidates, postdoctoral and senior researchers that work on a variety of topics in computational imaging. Regular seminars and meetings facilitate a stimulating and supportive environment with many opportunities for both scientific exchange and personal development. Furthermore, the PhD candidate will have the opportunity to visit other international research groups for exchange and collaboration, and will learn how to present and disseminate the results of their PhD project on the most important international conferences for the topical area, such as NeurIPS, SIAM Imaging Science, or ICIAM.

Tenure-track position in Stochastics

The Stochastics group of CWI aims to strengthen its research activities at the interface of rigorous probability and their applications to [classical and/or modern, data driven] algorithms, optimization and/or artificial intelligence. We are looking for talented individuals who are keen to develop their own innovative line of fundamental research within this theme.

Current research topics in the group are percolation theory, spatial statistics, rigorous scaling methods for large random systems, large-deviations theory, queueing theory, and application of these techniques to topics such as dynamic pricing logistics and computer/communication and energy networks. The candidate will explore collaboration opportunities within the group and/or other research groups at CWI such as Machine Learning or Optimization.

Management Ondersteuner Quantum & Society/QuSoft (M/V/X) 32 uur/0,8 fte

Het Quantum Delta NL (QDNL) programma is een groeifondsprogramma, gefinancierd door het Ministerie van Economische Zaken en Klimaat. Het programma heeft als doel om in Nederland een toonaangevend ecosysteem te realiseren op het gebied van quantumtechnologie. Binnen dit programma zijn drie Katalysator-programma’s en vier actielijnen. Eén van de actielijnen is de Quantum & Society actielijn, die zich als doel heeft gesteld er voor te zorgen dat de in het QDNL-programma ontwikkelde quantumtechnologie wordt ingezet ten goede van de maatschappij. Quantum & Society is een samenwerking tussen de universiteit van Leiden, de Universiteit van Amsterdam (UvA), de Technische Universiteit Delft en ECP, het platform voor de informatiesamenleving. Het CWI levert de coördinator voor deze actielijn en beheert de programmatische middelen.

QuSoft is het onderzoekscentrum voor Quantum software, een samenwerking van het CWI en de UvA. In dit verband werken bijna 80 mensen samen aan funderend onderzoek naar quantum-algoritmen, protocollen en -software. QuSoft is een belangrijke kennishub binnen het QDNL-programma, zowel in de uitvoering van onderdelen van het programma, als in de regie en coördinatie daarvan.