Suchergebnis: Katalogdaten im Herbstsemester 2014
|Robotics, Systems and Control Master|
|252-0535-00L||Machine Learning||W||6 KP||3V + 2U||J. M. Buhmann|
|Kurzbeschreibung||Machine learning algorithms provide analytical methods to search data sets for characteristic patterns. Typical tasks include the classification of data, function fitting and clustering, with applications in image and speech analysis, bioinformatics and exploratory data analysis. This course is accompanied by a practical machine learning projects.|
|Lernziel||Students will be familiarized with the most important concepts and algorithms for supervised and unsupervised learning; reinforce the statistics knowledge which is indispensible to solve modeling problems under uncertainty. Key concepts are the generalization ability of algorithms and systematic approaches to modeling and regularization. A machine learning project will provide an opportunity to test the machine learning algorithms on real world data.|
|Inhalt||The theory of fundamental machine learning concepts is presented in the lecture, and illustrated with relevant applications. Students can deepen their understanding by solving both pen-and-paper and programming exercises, where they implement and apply famous algorithms to real-world data.|
Topics covered in the lecture include:
- Bayesian theory of optimal decisions
- Maximum likelihood and Bayesian parameter inference
- Classification with discriminant functions: Perceptrons, Fisher's LDA and support vector machines (SVM)
- Ensemble methods: Bagging and Boosting
- Regression: least squares, ridge and LASSO penalization, non-linear regression and the bias-variance trade-off
- Non parametric density estimation: Parzen windows, nearest nieghbour
- Dimension reduction: principal component analysis (PCA) and beyond
|Skript||No lecture notes, but slides will be made available on the course webpage.|
|Literatur||C. Bishop. Pattern Recognition and Machine Learning. Springer 2007.|
R. Duda, P. Hart, and D. Stork. Pattern Classification. John Wiley &
Sons, second edition, 2001.
T. Hastie, R. Tibshirani, and J. Friedman. The Elements of Statistical
Learning: Data Mining, Inference and Prediction. Springer, 2001.
L. Wasserman. All of Statistics: A Concise Course in Statistical
Inference. Springer, 2004.
|Voraussetzungen / Besonderes||Solid basic knowledge in analysis, statistics and numerical methods for|
CSE. Experience in programming for solving the project tasks.
|252-1407-00L||Algorithmic Game Theory||W||7 KP||3V + 2U + 1A||P. Widmayer|
|Kurzbeschreibung||Game theory provides a formal model to study the behavior and interaction of self-interested users and programs in large-scale distributed computer systems without central control. The course discusses algorithmic aspects of game theory.|
|Lernziel||Learning the basic concepts of game theory and mechanism design, acquiring the computational paradigm of self-interested agents, and using these concepts in the computational and algorithmic setting.|
|Inhalt||The Internet is a typical example of a large-scale distributed computer system without central control, with users that are typically only interested in their own good. For instance, they are interested in getting high bandwidth for themselves, but don't care about others, and the same is true for computational load or download rates. Game theory provides a particularly well-suited model for the behaviour and interaction of such selfish users and programs. Classical game theory dates back to the 1930s and typically does not consider algorithmic aspects at all. Only a few years back, algorithms and game theory have been considered together, in an attempt to reconcile selfish behavior of independent agents with the common good.|
This course discusses algorithmic aspects of game-theoretic models, with a focus on recent algorithmic and mathematical developments. Rather than giving an overview of such developments, the course aims to study selected important topics in depth.
- Introduction to classical game theoretic concepts.
- Existence of stable solutions (equilibria), algorithms for computing equilibria, computational complexity.
- The cost difference between an optimum under central control and an equilibrium under selfish agents, known as the "price of anarchy".
- Auction-like mechanisms and algorithms that "direct" the actions of selfish agents into a certain desired equilibrium situation.
- Selected current research topics of Algorithmic Game Theory, such as Web-Search Based Keyword Auctions, or Information Cascading in Social Networks
|Skript||No lecture notes.|
|Literatur||"Algorithmic Game Theory", edited by N. Nisan, T. Roughgarden, E. Tardos, and V. Vazirani, Cambridge University Press, 2008; |
"Game Theory and Strategy", Philip D. Straffin, The Mathematical Association of America, 5th printing, 2004
Several copies of both books are available in the Computer Science library.
|Voraussetzungen / Besonderes||Audience: Although this is a Computer Science course, we encourage the participation from all students who are interested in this topic.|
Requirements: You should enjoy precise mathematical reasoning. You need to have passed a course on algorithms and complexity. No knowledge of game theory is required.
|252-5051-00L||Advanced Topics in Machine Learning||W||2 KP||2S||J. M. Buhmann, T. Hofmann, A. Krause|
|Kurzbeschreibung||In this seminar, recent papers of the pattern recognition and machine learning literature are presented and discussed. Possible topics cover statistical models in computer vision, graphical models and machine learning.|
|Lernziel||The seminar "Advanced Topics in Pattern Recognition" familiarizes students with recent developments in pattern recognition and machine learning. Original articles have to be presented and critically reviewed. The students will learn how to structure a scientific presentation in English which covers the key ideas of a scientific paper. An important goal of the seminar presentation is to summarize the essential ideas of the paper in sufficient depth while omitting details which are not essential for the understanding of the work. The presentation style will play an important role and should reach the level of professional scientific presentations.|
|Inhalt||The seminar will cover a number of recent papers which have emerged as important contributions to the pattern recognition and machine learning literature. The topics will vary from year to year but they are centered on methodological issues in machine learning like new learning algorithms, ensemble methods or new statistical models for machine learning applications. Frequently, papers are selected from computer vision or bioinformatics - two fields, which relies more and more on machine learning methodology and statistical models.|
|Literatur||The papers will be presented in the first session of the seminar.|
|151-0623-00L||ETH Zurich Distinguished Seminar in Robotics, Systems and Controls||W||1 KP||1S||B. Nelson|
|Kurzbeschreibung||This course consists of a series of seven lectures given by researchers who have distinguished themselves in the area of Robotics, Systems, and Controls.|
|Lernziel||Obtain an overview of various topics in Robotics, Systems, and Controls from leaders in the field. Please see Link for a list of upcoming lectures.|
|Inhalt||This course consists of a series of seven lectures given by researchers who have distinguished themselves in the area of Robotics, Systems, and Controls. MSc students in Robotics, Systems, and Controls are required to attend every lecture. Attendance will be monitored. If for some reason a student cannot attend one of the lectures, the student must select another ETH or University of Zurich seminar related to the field and submit a one page description of the seminar topic. Please see Link for a suggestion of other lectures.|
|Voraussetzungen / Besonderes||Students are required to attend all seven lectures to obtain credit. If a student must miss a lecture then attendance at a related special lecture will be accepted that is reported in a one page summary of the attended lecture. No exceptions to this rule are allowed.|
|263-5210-00L||Probabilistic Artificial Intelligence||W||4 KP||2V + 1U||A. Krause|
|Kurzbeschreibung||This course introduces core modeling techniques and algorithms from statistics, optimization, planning, and control and study applications in areas such as sensor networks, robotics, and the Internet.|
|Lernziel||How can we build systems that perform well in uncertain environments and unforeseen situations? How can we develop systems that exhibit "intelligent" behavior, without prescribing explicit rules? How can we build systems that learn from experience in order to improve their performance? We will study core modeling techniques and algorithms from statistics, optimization, planning, and control and study applications in areas such as sensor networks, robotics, and the Internet. The course is designed for upper-level undergraduate and graduate students.|
- Search (BFS, DFS, A*), constraint satisfaction and optimization
- Tutorial in logic (propositional, first-order)
- Bayesian Networks (models, exact and approximative inference, learning) - Temporal models (Hidden Markov Models, Dynamic Bayesian Networks)
- Probabilistic palnning (MDPs, POMPDPs)
- Reinforcement learning
- Combining logic and probability
|Voraussetzungen / Besonderes||Solid basic knowledge in statistics, algorithms and programming|
|151-0107-20L||High Performance Computing for Science and Engineering (HPCSE) I||W||4 KP||4G||P. Koumoutsakos, M. Troyer|
|Kurzbeschreibung||This course gives an introduction into algorithms and numerical methods for parallel computing for multi and many-core architectures and for applications from problems in science and engineering.|
|Lernziel||Introduction to HPC for scientists and engineers|
1. Parallel Computing Architectures
|Inhalt||Programming models and languages:|
1. C++ threading (2 weeks)
2. OpenMP (4 weeks)
3. MPI (5 weeks)
Computers and methods:
1. Hardware and architectures
3. Particles: N-body solvers
4. Fields: PDEs
5. Stochastics: Monte Carlo
Class notes, handouts
- Seite 1 von 1