[Opt-Net] Spring School "Optimization and Data Science" in Novi Sad, March 13-17, 2017

[Ambros Gleixner]

Dear colleagues,

We would like to announce that the COST Action TD1207 on "Mathematical
Optimization in Decision Support Systems for Efficient and Robust Energy
Networks", http://www.energy-opt.eu/, will be holding a

Spring School on Optimization and Data Science

Date: March 13-17, 2017
Location: Novi Sad, Serbia
Website: http://www.optdata-springschool.com

The training school will highlight the strong algorithmic connections
between the fields of optimization and machine learning and showcase
innovative applications at their interface. The goal is to equip
students both with a theoretical understanding and hands-on programming
experience.

Topics:
- Convex optimization in machine learning
- Mixed-integer programming and exact ML models
- Adaptive general regression and classification in the big data context
- Deep learning software packages: CNTK, Tensorflow, Theano, ...
- Applications in transportation, logistics, and healthcare
- Artificial neural networks in theory and practice
- Genetic algorithms in neural network optimization
- Traditional data analysis techniques: seriation, regression, frequent
itemsets, and their combinatorial structure
- Parallelization methods and their bottlenecks
- Optimization and cloud computing
- ...

Confirmed speakers:
- Tim Conrad, Freie Universität Berlin
- Maria Pia Fanti, Politecnico di Bari
- Robert Gower, Ecole Normale Superieure
- Utz-Uwe Haus, Cray EMEA Research Labs
- Daniel Hulme, SATALIA
- Philipp Hungerländer, Univ. Klagenfurt
- Filip Kulić, University of Novi Sad
- Milan Rapaić, University of Novi Sad
- ...


Registration is FREE and closes on Monday, February 20th, or when all
places are filled.  A limited amount of travel support is available.


We are looking forward to welcoming you in Novi Sad!

The main organizers:

Milena Petković, Željko Kanović, Zoran Jeličić
Faculty of Technical Sciences, University of Novi Sad

Ambros Gleixner
Zuse Institute Berlin