Max Planck Institute for Intelligent Systems
This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these template messages)(Learn how and when to remove this template message)
|Established||March 18, 2011|
|Address||Heisenbergstr. 3, 70569, Stuttgart, and Max-Planck-Ring 4, 72076 Tübingen|
The Max Planck Institute for Intelligent Systems (MPI-IS) exists since March 18, 2011. Its Stuttgart location (the former MPI for Metals Research) is in the process of scientific reorientation; a new institute is located in Tübingen. The MPI for Intelligent Systems belongs to the Max Planck Society, a German research institution strong in basic research.
The Max Planck Institute for Intelligent Systems approaches intelligent systems from the viewpoint of learning in systems with perception-action loops while these systems interact with natural, complex environments.
The institute focuses on establishing the scientific foundations of perception, action and learning through interdisciplinary and collaborative research across scales and systems. The focus is on constructing, modeling and analyzing synthetic and hybrid perception-action systems using learning and self-organization, both for handling sensory complexity and for building structural/physical complexity: to perceive and act in the world, methods of learning and inference will be crucial; to build up or grow physical structure, we need methods for creating self-organizing systems.
In the long run, the goal will be to develop a systems’ level understanding of how intelligent systems – of either technological, biological, or hybrid nature – can bootstrap themselves towards competence in their environments.
Perception, Learning, Action
An organizing principle for the institute is the concept of a perception-action loop. Learning the structure and optimized functionality of such feedback loops is critical for all autonomous systems. Thus, understanding the principles of perception-action loops in diverse instantiations of autonomous systems will be one common theme across the research groups of the institute.
Intelligent Systems Research across all scales
While pursuing research at both the macro and micro scales, a key novel focus of the institute is the integration of ideas across scales.
- Autonomous Motion (under interim management, Bernhard Schölkopf)
- Empirical Inference (Bernhard Schölkopf)
- Haptic Intelligence (Katherine J. Kuchenbecker)
- Perceiving Systems (Michael Black)
- Physical Intelligence (Metin Sitti)
- Modern Magnetic Systems (Gisela Schütz)
- Theory of Inhomogeneous Condensed Matter (Siegfried Dietrich)
- Autonomous Vision (Andreas Geiger)
- Autonomous Learning (Georg Martius)
- Dynamic Locomotion (Alexander Badri-Sprowitz)
- Embodied Vision (Jörg Stückler)
- Intelligent Control Systems (Sebastian Trimpe)
- Locomotion in Biorobotic and Somatic Systems (Ardian Jusufi)
- Micro, Nano, and Molecular Systems (Peer Fischer)
- Movement Generation and Control (Ludovic Righetti)
- Physics for Inference and Optimization (Caterina De Bacco)
- Probabilistic Numerics (Philipp Hennig)
- Rationality Enhancement (Falk Lieder)
- Statistical Learning Theory (Ulrike von Luxburg)
- Smart Nanoplasmonics (Laura Na Liu)
Center for Learning Systems (CLS)
On November 30, 2015, the Max Planck Society (MPS) and the ETH Zürich opened the Max Planck ETH Center for Learning Systems (CLS). The center's scientists want to understand what the principles of learning are, both in theory and in real machines. Among other goals they hope to develop robots which can act autonomously in an unknown, complex environment. An objective of this cooperation between the MPS and the ETH is to create scientific and staff synergies, and to ensure that European research in this field remains internationally competitive.
The Center for Learning Systems organizes joint events such as summer schools, workshops, retreats etc. Moreover, there are joint PhD- and Postdoc projects. The CLS is headed by directors Thomas Hofmann, ETH Zürich and Bernhard Schölkopf, MPI for Intelligent Systems. Professors and scientists from both institutions are members of the center.
History of the MPI for Metals Research
The Max Planck Institute for Metals Research (MPI-MF) was founded in Berlin-Neubabelsberg in 1920 as one of the first Kaiser Wilhelm Institutes. In 1934 it moved to Stuttgart and in 1949 it was renamed "Max-Planck-Institut für Metallforschung" (MPI-MF). In 2002, the Institute was fully established on the Max Planck Campus Stuttgart-Büsnau, being in close interaction with the Max Planck Institute for Solid State Research, as well as the University of Stuttgart and the Fraunhofer campus, which are close by.
In the early days of the MPI-MF its main research areas were physical metallurgy and metal physics. In the 70s, new areas, such as engineering ceramics, intermetallics and composites were added. During the first decade of the millennium, the MPI-MF once again shifted its research focus to include new materials, such as organics, soft matter, liquids and biomaterials and by implementing new analytical techniques and synthesis concepts.
At last the name "Metals Research" was only half the story. The limits between the classical subjects physics, chemistry and engineering sciences became blurred.
- International Journal of Materials Research 102 (2011) 7, Carl Hanser Verlag GmbH & Co.KG
- New research emphasis "Intelligent Systems" - Press Conference of the Max Planck Society and the Federal State of Baden-Württemberg on February 17, 2011
- Denkorte; Max-Planck-Gesellschaft und Kaiser-Wilhelm-Gesellschaft; Brüche und Kontinuitäten 1911-2011, pp. 330–339. Editors: Peter Gruss, Reinhard Rürup, Sandstein Verlag ISBN 978-3-942422-01-7 (2010)
- Max Planck Institute for Intelligent Systems
- MPI for Intelligent Systems in Tübingen
- Research Reports of the MPI for Intelligent Systems, Stuttgart Location (in German, Abstract and Citation in English)
- Research Reports of the MPI for Intelligent Systems, Tübingen Location (in German, Abstract and Citation in English)