Focus on:  Hide Contributions
Back to Conference View
Login

TRIUMF-Helmholtz Workshop on Scientific Computing

→ Europe/Berlin
DESY

DESY

Notkestr. 85, D-22607 Hamburg, Germany
Description

Scientific computing is a critical component of much of the work that takes place in particle physics at related subjects. In recent years, the amount of data produced across the globe has increased exponentially at research facilities and private businesses alike. With rapid advances in large-scale computing, big data, machine learning, and quantum computing, these technologies are beginning to have serious implications on how we do our work, and it is imperative that we remain part of this fast-changing field.  

The Helmholtz Association is partnering with TRIUMF, Canadian and German universities and selected businesses to host a second workshop on selected topics in scientific computing at DESY in Hamburg, on 16/17 September 2019, to develop further collaboration and to explore new tools in scientific computing.

Note that remote connections to all rooms are provided. You can find the relevant connection details for each session in the detailed description of each session (click on the session and select "Session details") or by clicking on the small folder icon in the top-right corner of each session.

Questions and support
  • Monday, 16 September
    • 09:00 → 12:30
      Plenary session 1: Welcome and keynotes Main auditorium

      Main auditorium

      DESY
      Main auditorium remote connection
      • 09:00
        Welcome addresses 30m
        Slides
      • 09:30
        Keynote: High Luminosity LHC - Computing Models and Impact of Quantum Computing 50m
        Speaker: Prof. Tommaso Boccali (INFN Sezione di Pisa ; Università di Pisa ; Scuola Normale Superiore di Pisa)
        Slides
      • 10:20
        Coffee break 25m
      • 10:45
        Keynote: Perspectives on the Future of Data Intensive Computing 50m
        Keynote: working group on data analytics
        Speakers: Florin Manaila (IBM), Oliver Oberst (IBM)
      • 11:35
        Keynote: Building the Bridge to Exascale: Applications and Opportunities for Nuclear Physics 50m
        Speaker: Jack Wells
    • 12:30 → 13:30
      Lunch break 1h Canteen

      Canteen

      DESY
    • 13:30 → 14:30
      Plenary session 2: Keynote:
      Main auditorium remote connection
      • 13:30
        Keynote: Lattice Gauge Theories within Quantum Technologies 50m
        Speaker: Dr Enrique Rico Ortega (Basque Country Univ UPV/EHU & Ikerbasque / Spain)
    • 14:30 → 18:00
      Data Analytics Seminar room 3

      Seminar room 3

      DESY
      SR 3 remote connection
      • 14:30
        Introduction 10m
        Speakers: Frank Schluenzen (DESY), Dr Wojciech Fedorko (TRIUMF)
        Slides
      • 14:40
        The Helmholtz Analytics Toolkit (HeAT), a Distributed Data Analysis Framework 30m
        Speaker: Daniel Coquelin (FZJ)
      • 15:10
        Deep Learning Applications in Water Cherenkov Neutrino Detectors 30m
        Speaker: Dr Nick Prouse (TRIUMF)
        Slides
      • 15:40
        Deep Learning Applications in Photon Science imaging applications 30m
        Speaker: Dr Philipp Heuser (DESY)
        Slides
      • 16:10
        Coffe Break 20m
      • 16:30
        Gaussian Process extrapolation of No-Core Shell Model Calculations 30m
        Speakers: Michael Gennari (University of Waterloo), Peter Gysbers (TRIUMF)
        Slides
      • 17:00
        Application of Machine Learning in CMS 30m
        Speaker: Dr Dirk Kruecker (DESY)
        Slides
      • 17:30
        Applications of Machine Learning in Accelerator Operation 30m
        Speaker: Dr Sergey Tomin (European XFEL)
        Slides
    • 14:30 → 18:00
      Large-scale computing Seminar room 4a

      Seminar room 4a

      DESY
      SR 4a remote connection
      • 14:30
        TRUMF Tier 1 Centre 25m
        Speaker: Di Qing (TRIUMF)
        Slides
      • 14:55
        Large scale computing infrastructure at DESY - current status and planning of Interdisciplinary Data Analysis Facility (IDAF) 25m
        Speakers: Christian Voss (DESY), Yves Kemp (DESY)
        Slides
      • 15:20
        HPC at DESY - experience and challenges 25m
        Speaker: Sergey Yakubov (DESY)
        Slides
      • 15:45
        R&D for Future High Throughput Computing @ GridKa 25m
        Speaker: Dr Manuel Giffels (Karlsruher Institut für Technologie (KIT))
        Slides
      • 16:10
        Coffee Break 20m
      • 16:30
        C++ developments for large scale computing 25m
        Speaker: Matthias Kretz (GSI)
        Slides
      • 16:55
        GPU accelerated ab initio nuclear structure calculations on ORNL Summit 25m
        Speaker: Petr Navratil (TRIUMF)
        Slides
      • 17:20
        Jupyter for Large Scale Computing 25m
        Slides
    • 14:30 → 18:00
      Quantum Computing Seminar room 1

      Seminar room 1

      DESY

      Notkestr. 85, D-22607 Hamburg, Germany
      slides
      SR 1 remote connection
      • 14:30
        Reports from the Labs 1h 40m
        • A lattice gauge theory testbed for quantum computing 25m
          A simplified model, based on U(1) lattice gauge theory in 2+1 dimensions, for use as a testbed for quantum computation is constructed. Focusing on the spectrum of low-lying states, energy calculations using the variational quantum eigensolver are implemented. Some results of running experiments with this model on IBM Q devices are presented.
          Speaker: Richard Woloshyn (TRIUMF)
        • Quantum Computing at DESY 25m
          We describe ongoing and possible applications of quantum computing at DESY in the areas of theoretical and experimental particle physics, astroparticle physics and photon science. A detailled example of a variational quantum computation on a superconducting qubit platform at Rigetti is presented using a hydrogen atom.
          Speaker: Dr Karl Jansen (NIC, DESY)
          Slides
        • Quantum Heuristic Algorithms for Hard Planning Problems from Aerospace Research 25m
          Abstract: Quantum heuristic algorithms do not have a proven advantage over classical algorithms. However, there are indications that these approaches might outperform classical approaches for certain applications. Moreover, they are believed to work without quantum error correction and are therefore amenable to early quantum computing devices.Hard combinatorial optimization problems as they occur in logistics or traffic management are highly relevant for society and business. Even minor improvements in the solution quality can have a enormous impact in terms of costs. We present our work on mapping and solving hard real world planning problems from aerospace research with quantum heuristic algorithms like the Quantum Approximate Optimization Algorithm (QAOA) and quantum annealing (QA). In particular, we discuss the choice of representative but small problem instances as well as the mapping of the original problem to a form compatible with the device and algorithm at hand. The latter includes various obstacles like the handling of constraints, the choice of algorithm parameters and compiling.
          Speaker: Dr Thorge Müller (DLR)
        • Quantum Variational Autoencoders and their applications 25m
          Generative models are among the most promising approaches toward understanding unlabelled data. They have a wide range of applications in structured prediction, molecular & material design, image analysis, speech synthesis, and computer vision. They pair with supervised learning models to help perform ML tasks when labeling data is expensive or labels are only available in a different domain. Quantum Boltzmann machine is a powerful generative model that can naturally be implemented on a quantum annealing device. However, the development of quantum-classical hybrid (QCH) algorithms is critical to deploy state-of-the-art computational models on current commercially available devices. A Quantum Variational Autoencoder (QVAE) is one such hybrid algorithm that consists of a latent generative process, formalized as a quantum or classical Boltzmann machine (QBM or BM). A quantum annealing processor is used for sampling from the Boltzmann prior distribution. The classical autoencoding structure is realized by a deep neural network, which allows inference to, and generating samples from, the latent space. We have successfully employed D-Wave quantum annealers as Boltzmann samplers to train end-to-end QVAE. The hybrid structure of QVAE allows us to deploy current quantum annealing devices in a QCH generative model with latent variables that achieves competitive performance on datasets such as MNIST.
          Speaker: Hossein Sadeghi (D-Wave)
      • 16:10
        Coffee break 20m
      • 16:30
        Round Table Introduction of each Participant 30m
        Introduce briefly the present activity and future goals, mention topics in which there is interested for collaboration
        Google Document to enter Participant Information
      • 17:00
        Discussion on common items 1h
    • 18:30 → 21:00
      Workshop dinner 2h 30m Bistro / Canteen

      Bistro / Canteen

      DESY
  • Tuesday, 17 September
    • 09:00 → 12:25
      Plenary session 3 Main auditorium

      Main auditorium

      DESY
      Main auditorium remote connection
      • 09:00
        Keynote: How to know, where to look - prioritising in computer vision 40m
        Speaker: Simone Frintrop (UHH)
      • 09:40
        Keynote: quantum computing in ion traps 40m
        Speaker: Ferdinand Schmidt-Kaler (FZJ)
      • 10:20
        Coffee break 25m
      • 10:45
        Keynote: working group one large-scale computing 40m
        Speaker: Tiziana Ferrari
      • 11:25
        Report: working group on quantum computing 15m
        Slides
      • 11:40
        Report: working group on large-scale computing 15m
        Slides
      • 11:55
        Report: working group on data analytics 15m
        Slides
    • 12:25 → 13:30
      Lunch break 1h 5m Canteen

      Canteen

      DESY
    • 13:30 → 15:10
      Plenary Session 4: Panel discussion on next steps and future plans Main auditorium

      Main auditorium

      DESY
      Inputfile for Questions and Comments
      Main auditorium remote connection
      slides
Powered by Indico v3.2.9