Technische Universität München, Zentrum Mathematik
Hauptseminar
Recent Advances in Nonlinear Optimization
Wintersemester 2020/2021
Content of the seminar
The seminar addresses recent advances in
Nonlinear Optimization and also serves as a preparation for a Bachelor's
or Master's thesis in this field. The seminar covers recent advances in the
following fields: Recent advances in optimization theory and methods,
especially also in data science and machine learning; novel applications
in technology, engineering, natural sciences, AI, etc.
Deutsche Version: Dieses Seminar widmet sich aktuellen Entwicklungen der Nichtlinearen
Optimierung und bereitet zudem auf eine Bachelor- oder Masterarbeit in
diesem Gebiet vor. In den Vorträgen werden insbesondere aktuelle
Themen aus folgenden Bereichen der nichtlinearen Optimierung
behandelt: neue Entwicklungen in Theorie und Methoden der Optimierung,
speziell auch in den Bereichen Data Science und Machine Learning;
innovative Anwendungen in Technik, Natur- und Wirtschaftswissenschaften,
KI, etc.
| Date |
|
Title |
| 10.12.20 |
|
Shape optimization using the method of mappings |
| 10.12.20 |
|
Existence and Optimality Conditions for Risk-Averse PDE Constrained Optimization |
| 17.12.20 |
|
Nonconvex optimization for machine learning: how to avoid saddle points |
| 17.12.20 |
|
Gossip algorithms for distributed learning |
| 07.01.21 |
|
BFGS method for problems with errors |
| 07.01.21 |
|
Derivative-free optimization with direct search methods |
| 14.01.21 |
|
Augmented Lagrangian methods |
| 14.01.21 |
|
Projection-free algorithms with applications in machine learning |
| 21.01.21 |
|
A majorization-minimization method for large-scale learning problems |
| 21.01.21 |
|
A trust-region method for Nash equilibrium problems |
| 28.01.21 |
|
Optimization problems with complementarity constraints |
| 28.01.21 |
|
Polynomial optimization and sums of squares |
| 04.02.21 |
|
Gradient sampling algorithm for nonsmooth optimization on manifolds |
| 04.02.21 |
|
Stochastic first-order methods with random projections |
Announcements
| 05.11.2020 |
A schedule of the talks is posted above. All talks will be held in English. The talks will take place from 2pm to 4pm. |
| 14.10.2020 |
A preliminary meeting will take place in the first week of the semester. |
Registration
Recommended prerequisites for the seminar:
For Bachelor students: Nichtlineare Optimierung: Grundlagen (MA2503); beneficial: (parallel visit of) Lineare und konvexe Optimierung (MA2504) or Nonlinear Optimization: Advanced (MA3503).
For Master students: Nichtlineare Optimierung: Grundlagen (MA2503), Nonlinear Optimization: Advanced (MA3503); beneficial: Lineare und konvexe Optimierung (MA2504)
Registration for the seminar was possible on TUMonline. More information on the registration process can be found on
http://www.ma.tum.de/Studium/HauptSeminareMaster.
