Deutsches Elektronen-Synchrotron - DESY

Ansprechpartner

Short summary of your group's research: We develop innovative methods to obtain full control over large molecules and nanoparticles. These methods and the created controlled samples are exploited in fundamental physics and chemistry studies to unravel the underlying mechanisms of chemistry and biology by watching molecules at work. This is coupled with advanced experimental control and data acquisition software, including on-the-fly data reduction, as well as theoretical and computational physics – both utilizing machine learning approaches for improved performance.

What infrastructure, programs and tools are used in your group? 

Have a look on our website.

What could a participant of the HIDA Trainee Network learn in your group? How could he or she support you in your group? Machine learning in theoretical physics and quantum chemistry and in real-time data reduction – benefit from current expertise and improve our approaches.

Ansprechpartner

Short summary of your group's research: We are developing computational tools for processing large crystallographic datasets consisting of many hundreds of thousands of individual frames.  Recently, we are moving towards real-time data processing, with the aim of making the data analysis step "disappear" - becoming part of the data acquisition process.  Our main product embodying this work is the free and open source "CrystFEL" software package.

What infrastructure, programs and tools are used in your group? 

Infrastructure: Gitlab, DESY "Maxwell" cluster

Tools: HDF5, SLURM, GSL, OpenCL, ZeroMQ, MessagePack, CMake, Meson, GTK and more

Languages: C/C++, Python, Lisp and others         

What could a participant of the HIDA Trainee Network learn in your group? How could he or she support you in your group? As well as working with large-scale accelerator-based photon facilities (PETRA III and European XFEL, plus other facilities worldwide), you would gain insight into developing a widely-used (>100 worldwide users) domain-specific software package.  For instance, how to balance the implementation of cutting-edge scientific methods with other considerations such as stability and reliability.  During your stay, you could contribute (for example) by experimenting with a new way of processing data, speeding up existing processing methods, or implementing one of the many requested features in CrystFEL.

Ansprechpartner

Short summary of your group's research: Our group is working on the development of computational techniques and high-accuracy simulations to improve our knowledge of the nuclear dynamics and spectra of molecules interacting with external electromagnetic fields. We're also working on ways to simulate ultrafast imaging experiments like photo-electron circular dichroism and laser-induced electron diffraction. We have recently begun to investigate various machine learning tools in applications to our problems, which range from solving differential equations to developing effective representations for molecular potential energy surfaces, as well as analyzing and inverting experimental photoelectron imaging data. 

What infrastructure, programs and tools are used in your group? HPC, scientific computing and machine learning tools implemented in Python and Julia.

What could a participant of the HIDA Trainee Network learn in your group? How could he or she support you in your group? You will develop or expand your grasp of many theoretical, computational, and experimental aspects of quantum molecular dynamics, high-resolution spectroscopy, photoelectron imaging, and machine learning applications to these challenges. You may learn more about software development techniques by contributing into development of open source software for molecular dynamics simulations.
We offer highly exploratory projects in development of neural networks for efficient solution of Schrödinger equation for nuclear motion and photo-electron dynamics, mathematical analysis of these approaches, and analysis of experimental photoelectron imaging data. We welcome applicants from a variety of disciplines with some background in machine learning and can offer and adapt a range of possible projects.

Ansprechpartner

Short summary of your group's research:

We are performing high energy neutrino astronomy using the observatories IceCube and RNO-G, as well as as the future IceCube-Gen2. Furthermore, we are working on optical surveys such as the Zwicky Transient Facility and the UV satellite ULTRASAT. The joint exploitation of the data enables new routes in multimessenger astrophysics. A key tool to combine the multimessenger data is AMPEL, a software framework developed by my research group to analyses realtime time series.

What infrastructure, programs and tools are used in your group?

AMPEL, IceCube, ZTF

What could a guest researcher learn in your group? How could he or she support you in your group?

Inference of galaxy properties and properties of astronomical transients using machine learning applications on multi band time series and catalog data. 

Ansprechpartner

Short summary of your group's research: 

Intelligent Process Controls (IPC) is a subgroup of the Machine Beam Controls (MSK) group at DESY, pushing forward innovative research for the autonomous operation of particle accelerators at the interface of machine learning, control. For this using reinforcement learning and other cutting-edge optimization techniques. IPC is also engaged in developing advanced feedbacks and enhancing fault diagnosis and anomaly detection through machine learning.

What infrastructure, programs and tools are used in your group?

The participant will have access to high-performance computing infrastructure at DESY. As programming languages, we mainly use Python.

What could a guest researcher learn in your group? How could he or she support you in your group?

As an interdisciplinary research team of experts from control theory, computer science and physics, a participant of the HIDA Trainee Network can gain experience in different directions and support many different applied projects. Here data mining projects as for anomaly detection are possible but also control and optimization problems. For the latter, we strongly encourage the Trainee to participate in shifts and applying the developed methods to the accelerator.