Seminars

Summer Term 2017

Entwicklung von Simulationen und wissenschaftlicher Software

Lecturer

JP Andreas Pott, Institute for Control Engineering of Machine Tools and Manufacturing Units 

Topics

Ausgewählte Aspekte der angewandten Entwicklung von Software für Simulationen, wissenschaftliche Berechnungen und Steuerungsfunktionen. Dies beinhaltet Methoden der Softwareentwicklung, Programmiertechniken, Entwicklungswerkzeuge und Aspekte des Projektmanagements. Typische Themen sind daher Spezifikation, Dokumentation, Qualitätssicherung, Design Pattern, Programmierparadigmen, SCRUM, Code-Generierung,  Lizensierung, Buildsysteme. Weitere Themen können von den Teilnehmern eingebracht werden.

Language

German

First Meeting

May, 12th, 2017, 15-17 pm, seminar room ISW 5.055

Time and place

Block seminar– dates by agreement, prospective in July

SWS

2

ECTS

3

Proof of attendance

Presentation and presence

Registration

via E-Mail to andreas.pott@isw.uni-stuttgart.de

User Centred Resear

Lecturer

Prof. Albrecht Schmidt, JP Niels Henze

Topics

EXPERIMENTAL RESEARCH
EXPERIMENTAL DESIGN
STATISTICAL ANALYSIS
SURVEYS
DIARIES
CASE STUDIES
INTERVIEWS AND FOCUS GROUPS
ETHNOGRAPHY
USABILITY TESTING
ANALYZING QUALITATIVE DATA
AUTOMATED DATA COLLECTION METHODS
MEASURING THE HUMAN
ONLINE RESEARCH
ETHICAL CONCERNS

Literature/Book: Jonathan Lazar, Jinjuan Heidi Feng, Harry Hochheiser (2011). Research Methods in Human-Computer Interaction. ISBN 0470723378 (kindle edition). New Version in 2017.

Evidence (Presentation, exam, presence): Presence, Summary, Presentation

Weblink: No

Preliminary discussion: first semester week

Regular cycle of the event: 3 dates per semester + block seminar

Language

English

First Meeting

fist week of summer term

Time and Place

3 courses during the term + block course

SWS

2

ECTS

3

Proof of attendance

presentation, summary and attendance

Registration

yes, albrecht.schmidt@vis.uni-stuttgart.de

"Stochastic and Statistical Topics in Modeling and Simulation" Part 3: Stochastic Approaches for Data-Integrated Simulation

Lecturer

Prof. Dr.-Ing. Wolfgang Nowak

Topics

Data-integrated simulation (DIS) refers to all attempts that seek to include
data more deeply into simulation. This ranges from purely data-driven machine learning and
statistical regression over data-driven model selection to data assimilation and data-driven
modelling of simulation errors. This seminar would like to provide an overview about the Bayesian/
statistical view onto DIS. The main focus will be on Bayesian regression, Bayesian model
selection and model averaging, error modelling, Bayesian parameter inference and data assimilation.
Classical machine learning (like artificial neural networks) are not in the focus, but
may be covered for the sake of the larger picture.

Language

German, English when required

Time and Place

Mondays, 11:30, starting on 8. May 2017 (distribution of topics to participants),
followed by first talks on 15. May 2017. The Seminar will run a bit into the semester
break to compensate for the late start.

Seminar room 2 (1.002) at the IWS (PWR61)

SWS

2

ECTS

3

Proof of attendance

Tutorial talks or key paper presentations (topics provided at first meeting), active
in-class discussion.

Registration

If you are interested, please send an email to Alica.Abberger@iws.uni-Stuttgart.de so that we
can anticipate the number of participants and distribute further information.

Modelling and Optimization of Hard Problems (BYOP)

Lecturer

Dr.  Colin Glass, HLRS

Topic
This block seminar will deal with hard problems. Modelling the problem is the basis for generating data and acquiring insight. The modelling process requires a good understanding of the field and of the goals. Starting from the model, simulations (or experiments) can be performed. The ensuing optimization step targets achieving the best result, whatever the contextual constraints may be.
 
For optimization methods, a close co-design with the specific problem is of paramount importance. The difference in performance easily amounts to orders of magnitude, depending on the suitability of the employed method. While there are some rules of thumb and some mappings of classes of problems to optimization methods, the task of developing an optimization method largely relies on expert insight both into the nature of the problem and into optimization techniques. Therefore, the field is intrinsically interdisciplinary and the degree of communication required is high. This block seminar aims at conveying sufficient understanding of the optimization domain in order to tackle quite hard problems and to establish a basis of understanding for efficient collaboration with optimization experts to approach hard problems.
 
All basic concepts from optimization, such as representation, evaluation function, fitness landscapes, constraints handling etc. will be covered. 
 
In the process of looking into modelling and optimization, hands-on intermezzi will be frequent, catering to any preferred form of collaborative effort between the participants and allowing to experience the potentially heavy impact of subtle changes in models or methods first hand. Please also Bring Your Own Problem (BYOP), which can be interlaced with the enfolding topics as appropriate and on an individual basis. A short description of your problem should be communicated in advance if possible.
 
After the block seminar, a short presentation either in the context of BYOP or examples from the seminar shall be given, adhering to conference standards regarding form (time, clarity, structure, etc.), while maintaining an amicable setting.
Language

The seminar will be held in English, except if all (!) participants prefer German.

Time and Place
Block seminar: 4 days + short presentation
 
Time: to be determined, communicate your availability. Planned: August ‘17
Contact: Colin W. Glass, glass(at)hlrs.de
 
HLRS, Nobelstraße 19
Proof of Attendance

Attendance and Presentation

SWS

2

ECTS

3

Literature

How to solve it: Modern Heuristics (Z. Michalewicz, D. Fogel)

Registration
Colin W. Glass, glass(at)hlrs.de
 

Contact

Barbara Teutsch
 

Barbara Teutsch

Coordinator Graduate School, Coach

To the top of the page