Petros Koumoutsakos: Catalogue data in Spring Semester 2018

Name	Dr. Petros Koumoutsakos
URL	http://www.cse-lab.ethz.ch/index.php?&option=com_content&view=article&id=100&catid=38
Department	Mechanical and Process Engineering
Relationship	Full Professor

Number	Title	ECTS	Hours	Lecturers
101-0190-08L	Uncertainty Quantification and Data Analysis in Applied Sciences The course should be open to doctoral students from within ETH and UZH who work in the field of Computational Science. External graduate students and other auditors will be allowed by permission of the instructors.	3 credits	4G	E. Chatzi, P. Chatzidoukas, P. Koumoutsakos, S. Marelli, V. Ntertimanis, K. Papadimitriou, B. Sudret
Abstract	The course presents fundamental concepts and advanced methodologies for handling and interpreting data in relation with models. It elaborates on methods and tools for identifying, quantifying and propagating uncertainty through models of systems with applications in various fields of Engineering and Applied science.
Learning objective	The course is offered as part of the Computational Science Zurich (CSZ) (http://www.zhcs.ch/) graduate program, a joint initiative between ETH Zürich and University of Zürich. This CSZ Block Course aims at providing a graduate level introduction into probabilistic modeling and identification of engineering systems. Along with fundamentals of probabilistic and dynamic system analysis, advanced methods and tools will be introduced for surrogate and reduced order models, sensitivity and failure analysis, parallel processing, uncertainty quantification and propagation, system identification, nonlinear and non-stationary system analysis.
Content	The topics to be covered are in three broad categories, with a detailed outline available online (see Learning Materials). Track 1: Uncertainty Quantification and Rare Event Estimation in Engineering, offered by the Chair of Risk, Safety and Uncertainty Quantification, ETH Zurich (20 hours) Lecturers: Prof. Dr. Bruno Sudret, Dr. Stefano Marelli Track 2: Bayesian Inference and Uncertainty Propagation, offered the by the System Dynamics Laboratory, University of Thessaly, and the Chair of Computational Science, ETH Zurich (20 hours) Lecturers: Prof. Dr. Costas Papadimitriou, Dr. Panagiotis Hadjidoukas, Prof. Dr. Petros Koumoutsakos Track 3: Data-driven Identification and Simulation of Dynamic Systems, offered the by the Chair of Structural Mechanics, ETH Zurich (20 hours) Lecturers: Prof. Dr. Eleni Chatzi, Dr. Vasilis Dertimanis. The lectures will be complemented via a comprehensive series of interactive Tutorials will take place.
Lecture notes	The course script is composed by the lecture slides, which will be continuously updated throughout the duration of the course on the CSZ website.
Literature	Suggested Reading: Track 2 : E.T. Jaynes: Probability Theory: The logic of Science Track 3: T. Söderström and P. Stoica: System Identification, Prentice Hall International, Link see Learning Materials. Xiu, D. (2010) Numerical methods for stochastic computations - A spectral method approach, Princeton University press. Smith, R. (2014) Uncertainty Quantification: Theory, Implementation and Applications SIAM Computational Science and Engineering, Lemaire, M. (2009) Structural reliability, Wiley. Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., Saisana, M. & Tarantola, S. (2008) Global Sensitivity Analysis - The Primer, Wiley.
Prerequisites / Notice	Introductory course on probability theory Fair command on Matlab
151-0116-00L	High Performance Computing for Science and Engineering (HPCSE) for CSE	7 credits	4G + 2P	P. Koumoutsakos, P. Chatzidoukas
Abstract	This course focuses on programming methods and tools for parallel computing on multi and many-core architectures. Emphasis will be placed on practical and computational aspects of Bayesian Uncertainty Quantification and Machine Learning including the implementation of these algorithms on HPC architectures.
Learning objective	The course will teach - programming models and tools for multi and many-core architectures - fundamental concepts of Uncertainty Quantification and Propagation (UQ+P) for computational models of systems in Engineering and Life Sciences. - fundamentals of Deep Learning
Content	High Performance Computing: - Advanced topics in shared-memory programming - Advanced topics in MPI - GPU architectures and CUDA programming Uncertainty Quantification: - Uncertainty quantification under parametric and non-parametric modeling uncertainty - Bayesian inference with model class assessment - Markov Chain Monte Carlo simulation Machine Learning - Deep Neural Networks and Stochastic Gradient Descent - Deep Neural Networks for Data Compression (Autoencoders) - Recurrent Neural Networks
Lecture notes	http://www.cse-lab.ethz.ch/index.php/teaching/42-teaching/classes/704-hpcse2 Class notes, handouts
Literature	- Class notes - Introduction to High Performance Computing for Scientists and Engineers, G. Hager and G. Wellein - CUDA by example, J. Sanders and E. Kandrot - Data Analysis: A Bayesian Tutorial, Devinderjit Sivia - Machine Learning: A Bayesian and Optimization Perspective, S. Theodorides
Prerequisites / Notice	Attendance of HPCSE I
151-0116-10L	High Performance Computing for Science and Engineering (HPCSE) for Engineers II	4 credits	4G	P. Koumoutsakos, P. Chatzidoukas
Abstract	This course focuses on programming methods and tools for parallel computing on multi and many-core architectures. Emphasis will be placed on practical and computational aspects of Uncertainty Quantification and Propagation including the implementation of relevant algorithms on HPC architectures.
Learning objective	The course will teach - programming models and tools for multi and many-core architectures - fundamental concepts of Uncertainty Quantification and Propagation (UQ+P) for computational models of systems in Engineering and Life Sciences
Content	High Performance Computing: - Advanced topics in shared-memory programming - Advanced topics in MPI - GPU architectures and CUDA programming Uncertainty Quantification: - Uncertainty quantification under parametric and non-parametric modeling uncertainty - Bayesian inference with model class assessment - Markov Chain Monte Carlo simulation
Lecture notes	http://www.cse-lab.ethz.ch/index.php/teaching/42-teaching/classes/704-hpcse2 Class notes, handouts
Literature	- Class notes - Introduction to High Performance Computing for Scientists and Engineers, G. Hager and G. Wellein - CUDA by example, J. Sanders and E. Kandrot - Data Analysis: A Bayesian Tutorial, Devinderjit Sivia
151-0431-00L	Models, Algorithms and Data: Introduction to Computing	4 credits	2V + 1U	P. Koumoutsakos, J. H. Walther
Abstract	Fundamental Computational Methods for data analysis, modeling and simulation relevant to Engineering applications. The course emphasizes the implementation of these methods using object oriented programming in C++ with application examples drawn from Engineering applications
Learning objective	The course aims to introduce Engineering students to fundamentals of Interpolation, Solution of non-linear equations, Filtering and Numerical Integration. The course aims to integrate numerical methods with enhancing the students programming skills in object oriented languages. The course serves as foundation for Computational Methods in Engineering Applications II (Fall Semester), that is concerned with Ordinary and Partial Differential Equations.
Lecture notes	Lecture Notes will be distributed in class
Literature	1. Introduction to Applied Mathematics, G. Strang 2. Analysis of Numerical Methods, Isaacson and Keller
Prerequisites / Notice	- Informatik - A course on the interface of classical (first principle) and Data driven models in computing. Fundamental algorithms for inference, approximation and optimisation. Bridging the gap of Computational and Data sciences.