Dr. Guido Montúfar, research group leader at the Max Planck Institute for Mathematics in the Sciences Leipzig, has been awarded an ERC Starting Grant for excellent young researchers. The European Research Council ERC is supporting his new group on deep learning theory, which focuses on improving learning in neural networks, with an amount of 1.5 million euro.
Deep Learning is one of the most vibrant areas of contemporary machine learning and one of the most promising approaches to Artificial Intelligence. This research area drives the latest systems for image, text, and audio processing, as well as an increasing number of new technologies. The goal of this new research group is to advance on key open problems in Deep Learning, specifically those regarding the capacity, optimisation, and regularisation of the underlying algorithms.
The scientists will consolidate a theoretical basis that allows to pin down the inner workings of the present success of Deep Learning and make it more widely applicable, in particular in situations with limited data and challenging problems in reinforcement learning. The scientific approach is based on the geometry of neural networks and exploits innovative mathematics, drawing on information geometry and algebraic statistics.
Guido Montúfar studied mathematics and physics at the Technical University of Berlin and received his doctorate at the University of Leipzig. After working as a research associate at the Pennsylvania State University, he obtained a postdoc position at the Max Planck Institute for Mathematics in the Sciences.
In parallel to his recent position as research group leader he holds an Assistant Professorship at the Departments of Mathematics and Statistics at the University of California, Los Angeles, USA. With his team at UCLA, he develops mathematical tools and techniques for computation with neural networks, with diverse applications from generative modelling, optimisation, to pure mathematics.
Guido Montúfar’s work is dedicated to advancing on the most important challenges in deep learning today, with deep and direct practical relevance. Together with his team, he pursues a synergistic approach at the intersection of Mathematics, Statistics, Machine learning, merging tools from information theory, algebra, combinatorics, and geometry.
The geometric analysis of deep neural networks that Guido has been developing over the past years provides a formal approach to the design of learning systems that allows, for instance, to create sparse networks with guaranteed learning capabilities. His work on neural networks includes the analysis of distributed representations, the advantages of depth in function approximation, the geometry of graphical models with hidden variables.
This work is opening up new avenues for addressing one of the most serious bottlenecks in contemporary reinforcement learning, namely the sample complexity of these methods. This work is also facilitating the development of new optimization algorithms for learning with neural networks and new regularization techniques based on information theory.
Guido is also an active coordinator of research activities targeting interactions between mathematics and machine learning, such as the recent DALI workshop on “Theory of Deep Learning” and the upcoming AIM workshop “Boltzmann Machines” at the American Institute of Mathematics. Guido Montúfar’s team is also closely involved in the Mathematics of Data initiative at the Max Planck Institute for Mathematics in the Sciences.
ERC Starting Grants are awarded by the European Research Council to excellent junior researchers who have already achieved outstanding research results. The funding is used to set up an own working group.
Dr. Guido Montúfar
Max Planck Institute for Mathematics in the Sciences
http://www.math.ucla.edu/~montufar Information about the laureate Dr. Guido Montúfar:
https://erc.europa.eu/funding/starting-grants Information about the ERC Starting Grants Program
http://www.mis.mpg.de/math-of-data Information about the „Mathematics of Data“-Initiative at the Max Planck Institute for Mathematics in the Sciences
http://www.mis.mpg.de Information about the Max Planck Institute for Mathematics in the Sciences
Jana Gregor | Max-Planck-Institut für Mathematik in den Naturwissenschaften (MPIMIS)
MaterialVital Preis 2019 awarded for novel hydrogel wound dressings
05.09.2019 | Leibniz-Institut für Polymerforschung Dresden e. V.
Decoding cell communication
13.06.2019 | Friedrich-Alexander-Universität Erlangen-Nürnberg
Researchers from the Department of Atomically Resolved Dynamics of the Max Planck Institute for the Structure and Dynamics of Matter (MPSD) at the Center for Free-Electron Laser Science in Hamburg, the University of Hamburg and the European Molecular Biology Laboratory (EMBL) outstation in the city have developed a new method to watch biomolecules at work. This method dramatically simplifies starting enzymatic reactions by mixing a cocktail of small amounts of liquids with protein crystals. Determination of the protein structures at different times after mixing can be assembled into a time-lapse sequence that shows the molecular foundations of biology.
The functions of biomolecules are determined by their motions and structural changes. Yet it is a formidable challenge to understand these dynamic motions.
At the International Symposium on Automotive Lighting 2019 (ISAL) in Darmstadt from September 23 to 25, 2019, the Fraunhofer Institute for Organic Electronics, Electron Beam and Plasma Technology FEP, a provider of research and development services in the field of organic electronics, will present OLED light strips of any length with additional functionalities for the first time at booth no. 37.
Almost everyone is familiar with light strips for interior design. LED strips are available by the metre in DIY stores around the corner and are just as often...
Later during this century, around 2060, a paradigm shift in global energy consumption is expected: we will spend more energy for cooling than for heating....
Researchers from the Department of Atomically Resolved Dynamics of the Max Planck Institute for the Structure and Dynamics of Matter (MPSD) at the Center for Free-Electron Laser Science in Hamburg, the University of Potsdam (both in Germany) and the University of Toronto (Canada) have pieced together a detailed time-lapse movie revealing all the major steps during the catalytic cycle of an enzyme. Surprisingly, the communication between the protein units is accomplished via a water-network akin to a string telephone. This communication is aligned with a ‘breathing’ motion, that is the expansion and contraction of the protein.
This time-lapse sequence of structures reveals dynamic motions as a fundamental element in the molecular foundations of biology.
Two research teams have succeeded simultaneously in measuring the long-sought Thorium nuclear transition, which enables extremely precise nuclear clocks. TU Wien (Vienna) is part of both teams.
If you want to build the most accurate clock in the world, you need something that "ticks" very fast and extremely precise. In an atomic clock, electrons are...
10.09.2019 | Event News
04.09.2019 | Event News
29.08.2019 | Event News
18.09.2019 | Innovative Products
18.09.2019 | Physics and Astronomy
18.09.2019 | Materials Sciences