Program Organization

While quantum mechanics is crucial to understanding the properties of all materials, the term quantum materials has been introduced to refer to materials “with exotic physical properties, arising from the quantum mechanical properties of their constituent electrons”. Quantum materials challenge the established view of condensed matter in at least two ways. Firstly, strong interactions between electrons can lead to the emergence of conceptually new states, and secondly, non-trivial topologies of the wave functions and fields can guarantee exceptional stability of the relevant states. Both properties are clearly of major interest with respect to tailoring functionalities for new computing paradigms.

The strategic vision of the Topic research is to explore the fundamentals of quantum materials right down to the atomic and molecular scales, to engineer their properties with atomic precision, and, on this basis, to develop material platforms which lend themselves to innovative NC and QC device concepts.

Speaker of the Topic: Matthieu Le Tacon (KIT)

Quantum information processing promises exponential acceleration for a range of important computational tasks and for physically secure communication channels. In this Topic, quantum scientists and engineers combine their complementary expertise to scale up the state of the art in information processing devices, develop advanced qubit control electronics and explore new physical realizations of qubits. The overarching goal is to demonstrate the quantum advantage in real-world applications.

Speaker of the Topic: David DiVincenzo (FZJ)

The investigation of universal principles in the brain and their exploitation for computation, requires convergence between multiple scientific disciplines. Brain activity is observed using electro- and optophysiological methods and the findings are interpreted with the help of advances in time-resolved statistical methods for massively parallel data, network modeling, simulation technology, and analytical theory.

The objective is to construct modular computing devices that combine the best of a supercomputer with von Neumann architecture, implementing traditional algorithms and providing high performance for floating point calculations, enhanced by neuromorphic accelerators that employ biological principles of information processing. This approach will minimize the energy to solution by choosing methods that are adapted to the nature of the problem. All levels of neuromorphic computing are addressed from materials, over circuit design and system integration to community building.

Speaker of the topic: Markus Diesmann (FZJ)

Life is an information processing process. To avoid the pitfalls, and exploit the opportunities of a dynamic environment, organisms must interpret a manifold of physical and chemical signals. These signals represent information of vital importance whose processing is essential to ensure appropriate reactions. Modern science has dissected many of the elementary steps of biological information processing. What has emerged is a breathtaking picture: a plethora of cells and molecules interacting in a massively parallel and intertwined way. While each process appears noisy and slow, overall organisms are stunningly efficient and successful information processors. Via a multidisciplinary approach and by shifting the focus towards an integration of such elementary processes into an overall information processing picture, Topic 4 aims at a paradigm shift in the life sciences with far reaching consequences for our understanding of life that will enable new bioinspired technologies.

Speaker of the topic: Andreas Offenhäusser (FZJ)

Topic 5 addresses the Connectome as a multilevel system underlying the brain’s function, dysfunction and plasticity, analyses the principles of information processing, and applies such insights for developing novel diagnostics, model-based therapeutics and new methods, e.g., machine/deep learning.

Neuroimaging techniques and optical methods shed light on brain structure and function at multiple spatial and temporal scales, while considering intersubject variability. Empirical research is supplemented by mathematical and data-driven modelling and simulation. The multimodal brain atlas provides a unique framework to integrate the different levels of analysis, and to link the different pieces of research. The challenge of analyzing large and complex data in HPC-based workflows is research target in our Joint Lab with Program 1. Researchers of Topic 5 are sharing tools, models and data to the international scientific community and co-develop research platforms.

Speaker of the topic: Katrin Amunts (FZJ)