This interactive tutorial has been created by students of CSE in the last ten months. http://run.coplon.de

One of the future challenges of CSE:

US company new “TUM Partner of Excellence” Google to invest in science "made in Germany"

Yesterday we celebrated this year's CSE graduates. Special thanks go to our student speakers giving a great speech and everybody else who participated!

We are happy to announce the release of v1.0.0 of preCICE. The development of tools like preCICE is a possible field of work for students of CSE. preCICE (Precise Code Interaction Coupling Environment) is a coupling library for partitioned multi-physics simulations, including, but not restricted to fluid-structure interaction and conjugate heat transfer simulations. preCICE is developed at the Chair of Scientific Computing at the Technical University of Munich and at the Institute for Parallel and Distributed Systems at the University of Stuttgart. https://github.com/precice/precice/releases/tag/v1.0.0

Ravi Kishore is a CSE '12 and BGCE '13 graduate currently doing his PhD in Mechanical Engineering and Computation at the Massachusetts Institute of Technology. He works on developing a mechanistic model to simulate flow boiling heat transfer, which allows one to design compact thermal systems that achieve very high local heat transfer rates. Current understanding of flow boiling is limited to empirical correlations and cannot be applied consistently to perform detailed simulations. His work focuses on a new bottom-up validation approach that can represent the underlying physics faithfully and ensures wide applicability. Ravi collaborates with an interdisciplinary team of researchers including experimentalists, physicists, engineers, applied mathematicians and computer scientists. The picture shows the flow boiling experimental loop at MIT, built to provide high resolution information to support the mathematical models.

Adam Kosiorek, CSE '14 graduate, is currently doing his PhD at the University of Oxford. His research focuses on attention mechanisms and external memory in recurrent neural networks for reinforcement learning. Humans often tackle problems by dividing them into subproblems, and then solving them one at a time. Attention mechanisms allow to address parts of a problem at a time, and external memory can further facilitate this task. Adam's long interest in neural networks increased during his Master's thesis, where he used Bayesian Recurrent Neural Networks to model time series of joint angles describing the movement of a human arm in time and space.

We are proud to announce that our colleagues and former CSE students Paul Cristian Sarbu and Ionut Farcas have won this year's "Best Poster Award" at the SIAM CSE conference in Atlanta! Their work introduces a new multilevel stochastic collocation method using dimension reduction, comparing different adaptive sparse grid strategies applied to dampened oscillators and fluid-structure interaction scenarios. Congratulations! #SIAMCSE17

Tonight we celebrated the CSE graduation ceremony. Congratulations to our CSE graduates! More photos coming soon :)