PhD student Scientific Computing with SDG
Four-year PhD position, position 2: TU Eindhoven (Netherlands) with SDG Milan (Italy)Two 4-year Marie Curie PhD positions are available at TU Eindhoven, …
- de Rondom, Eindhoven, Noord-Brabant
- Tijdelijk contract / Tijdelijke opdracht
- Uren per week:
- 38 uur
FunctieomschrijvingFour-year PhD position, position 2: TU Eindhoven (Netherlands) with SDG Milan (Italy)
Two 4-year Marie Curie PhD positions are available at TU Eindhoven, starting date March 1, 2019. Research topics may include numerical linear algebra, big data, large-scale optimization, modeling, model reduction, probability and statistics.
This is a description of Position 2 at TU Eindhoven, joint with SDG Milan, www.sdggroup.com/en. This position is 'ESR6' (Early-stage researcher), part of the EU Marie Curie EID (European Industrial Doctorate) project BIGMATH, itn-bigmath.unimi.it , including 7 PhD positions in total, at universities in Milan, Novi Sad, Lisbon, and Eindhoven.
You will be a member of the Centre for Analysis, Scientific Computing and Applications (CASA), within the Department of Mathematics and Computer Science at TU Eindhoven. Your daily advisors will be:
Moreover, at TU Eindhoven, you may also work with
Additionally, you will spend 2 months at the University of Milan and work with
Keywords: Data science, numerical linear algebra, algebraic modeling, predictive modeling, predictive manufacturing, data reduction, model reduction, optimization, statistics, machine learning.
Please find below a brief description from the BIGMATH proposal. Here 'RO' stands for research objective. This position is ESR6 (ESR = Early-stage researcher).
Project 6 (ESR6): Prediction of failure events in complex productive systems (TU/e & SDG)
With the advent of Industry 4.0, many current industrial processes are subject to continuous monitoring of their efficiency by sensors, which provide numerical data of various types, with a high frequency. The data gathered by these systems have various uses. Two important but difficult objectives are (1) prediction of events that can impair (damage) the process output, and (2) prediction and optimization of the quality of the process end-product. However, the availability of vast amount of data leads business users to even more ambitious objectives: (3) understanding the causes of failures and varying quality, and (4) taking actions to improve the process, to avoid failures or to reduce the failure rates. These goals are very challenging. Good and reliable prediction usually requires both linear and nonlinear models and algorithms, which are often difficult to interpret, so that causal relationships remain unclear. On the contrary, interpretable models often are unable to represent complex, nonlinear and dynamic relationships. Therefore, ESR6 will develop efficient models that are able to predict with a good reliability the occurrence of process impairment, and to tell which features mostly contribute to the damage in the given conditions. Three key mathematical ingredients that we will use are: (a) compositionality: building complex models from simpler components; (b) model reduction: simplification of the model, so that it can be simulated and the parameters identified, without losing much accuracy (RO3); (c) causality: understanding what is causing what and take improvement-oriented actions (RO4).
Current trends for impairment prediction frequently use deep learning models, which are flexible and often provide good results, but usually very hard to interpret. In this project, ESR6 will introduce
innovative more interpretable models. Such models will support the development of innovative methods for process optimization and control by ESR7.
The relevant research objectives:
RO3: Develop model reduction or feature selection techniques for the construction of fit-for-purpose
models, which may reduce the complexity of a system, increasing the interpretability of cause-effect
RO4: Develop interpretable statistical models for classification in imbalanced classes and for the
prediction of rare events (i.e. classification into 2 imbalanced classes). The aim is to overcome the
application of 'black box' machine learning techniques, using models that can interpret the
interrelationships and the causal effects among different features.
Some information on SDG
SDG is a global consulting firm with a broad focus on data analytics in all its aspects. The data science practice is firmly rooted in strong mathematical and statistical know-how and a sound computing expertise. SDG is constantly growing by hiring many young people coming directly from universities, and some more experienced people to support the growth in a consistent way.
Locations: Eindhoven (Netherlands, 28 months) and Milan (Italy, 20 months), as follows:
March 2019: TU Eindhoven (with 1 week course in Milan); April—August 2019: SDG Milan; September 2019—August 2020: TU Eindhoven; September—October 2020: University of Milan; November 2020— November 2021: SDG Milan; December 2021—February 2023: TU Eindhoven.
Additional informationTo apply:
For informal inquiries, please contact Dr. Michiel Hochstenbach, TU Eindhoven by email. To apply, please use the TU Eindhoven system. by using the "apply now" button. Please include all of this: motivation letter, CV (math interests, languages, some personal info, hobbies), list of BSc and MSc courses with grades, MSc thesis (or draft), list of ca 3 people for recommendation. See also www.win.tue.nl/~hochsten/bigmath.html.
Scanning of the applications will start immediately; applications received before or on December 5, 2018 will receive full consideration, but the call will be open until the positions are filled.
You are also very welcome to obtain informal information about the project and SDG via Maurizio Sanarico, firstname.lastname@example.org.
For further information about employment conditions you may contact Marjolein von Reth. HR advisor TU Eindhoven, email@example.com.