Zoom Link
The KIT Graduate School Computational and Data Science (KCDS) at KIT Center MathSEE proudly presents: KCDS Talks, a monthly series of short lectures from basic knowledge to trending topics in computational and data science.
 
In July, KCDS Fellows present their research in:
1. Neural Nets for Solving Economic Models (Lukas Frank, ECON)
2. Data driven model for weather forecasting (Deifilia To, SCC/IMKTRO)
3. Reconstruction of Particle Position and Size in Dispersed Multiphase Flows using Deep Learning and Physics-Based Optimisation (Christian Sax, ISTM/IANM)
 
1. Neural Nets for Solving Economic Models (Lukas Frank, ECON)
Solving rich economic models globally often requires to solve a high-dimensional functional equation. With classical grid-based methods, the curse of dimensionality limits the number of model features to d ≈ 20. Recent advances leverage the abilities of neural nets to mitigate the curse of dimensionality, bringing more realistic models in reach. However, neural nets pose new challenges such as mediocre accuracy and fragile convergence behavior. I show how to solve high-dimensional economic models with neural nets and how to cure some of the most salient issues.
 
2. Data driven model for weather forecasting (Deifilia To, SCC/IMKTRO)
Traditional methods for weather forecasting are based on the solution of physical conservation equations that are grounded in theory. In contrast, current machine learning methods learn only through data. Machine learning methods can now create better forecasts than traditional methods - but their success is not well understood. I replicate and study one of the most well-known models, Pangu-Weather, and propose improvements in the architecture that could lead to more efficient training and accurate weather forecasts.
 
3. Reconstruction of Particle Position and Size in Dispersed Multiphase Flows using Deep Learning and Physics-Based Optimisation (Christian Sax, ISTM/IANM)
Dispersed multiphase flows play an important role in a multitude of environmental and industrial applications, such as spray, mist, cavitation and boiling. A novel diagnostic tool is developed for the investigation of such flows from single camera images. The approach combines deep learning for image segmentation and classification with the optimization of a non-linear functional incorporating a model of the scattering process.
 
If you are a master student, a doctoral researcher, a senior researcher or just interested in the topics - join us!
 
(for free and without registration)

During data assimilation, observations are optimally combined with a (numerical) model to obtain the best estimate of the system's state taking all observations into account. As observations are often noisy, incomplete and inconsistent, substantial observation pre-processing is required. In this session, we provide insight into various data pre-processing and visualization methods and introduce the concept of data assimilation using the example of numerical weather prediction. The practical exercises based on Jupyter notebooks will illustrate different methods for data processing and data assimilation.
 
The workshop will run from September 11 - 12 on-site at KIT Campus South in Karlsruhe and is open to (doctoral) researchers from KIT as well as other universities and research centers.
 
Find more information and apply to join the workshop

During data assimilation, observations are optimally combined with a (numerical) model to obtain the best estimate of the system's state taking all observations into account. As observations are often noisy, incomplete and inconsistent, substantial observation pre-processing is required. In this session, we provide insight into various data pre-processing and visualization methods and introduce the concept of data assimilation using the example of numerical weather prediction. The practical exercises based on Jupyter notebooks will illustrate different methods for data processing and data assimilation.
 
The workshop will run from September 11 - 12 on-site at KIT Campus South in Karlsruhe and is open to (doctoral) researchers from KIT as well as other universities and research centers.
 
Find more information and apply to join the workshop

Zoom Link
The KIT Graduate School Computational and Data Science (KCDS) at KIT Center MathSEE proudly presents: KCDS Talks, a monthly series of short lectures from basic knowledge to trending topics in computational and data science.
 
In July, Dr. Rebekka Buse (Statistical Methods & Econometrics at KIT) joins us for a talk entitled ""Econometric Methods For Dynamic Networks".
 
If you are a master student, a doctoral researcher, a senior researcher or just interested in the topics - join us!
 
(for free and without registration)

Accurately estimating rainfall by radar data is challenging because radars measure reflectivity rather than direct rainfall, and environmental variations further complicate this conversion. The RainQuest hackathon aims to address this problem by developing models that integrate precise point measurements from rain gauge data with radar reflectivity, which offers better measurements resolution. By combining these data sources, we aim to enhance the precision of precipitation estimates.
We invite all data science and machine learning enthusiasts to join us for this exciting challenge in a relaxed, collaborative environment. Participants will enhance their skills in data analysis, machine learning, and meteorological modeling. No prior experience with weather data is required.  Additionally, you will have the opportunity to connect with like-minded individuals and work with KIT´s supercomputer cluster. 
This event is organized by machine learning enthusiasts from KCDS, with support from MathSEE, TRIANGEL and SCC.

Accurately estimating rainfall by radar data is challenging because radars measure reflectivity rather than direct rainfall, and environmental variations further complicate this conversion. The RainQuest hackathon aims to address this problem by developing models that integrate precise point measurements from rain gauge data with radar reflectivity, which offers better measurements resolution. By combining these data sources, we aim to enhance the precision of precipitation estimates.
We invite all data science and machine learning enthusiasts to join us for this exciting challenge in a relaxed, collaborative environment. Participants will enhance their skills in data analysis, machine learning, and meteorological modeling. No prior experience with weather data is required.  Additionally, you will have the opportunity to connect with like-minded individuals and work with KIT´s supercomputer cluster. 
This event is organized by machine learning enthusiasts from KCDS, with support from MathSEE, TRIANGEL and SCC.

Accurately estimating rainfall by radar data is challenging because radars measure reflectivity rather than direct rainfall, and environmental variations further complicate this conversion. The RainQuest hackathon aims to address this problem by developing models that integrate precise point measurements from rain gauge data with radar reflectivity, which offers better measurements resolution. By combining these data sources, we aim to enhance the precision of precipitation estimates.
We invite all data science and machine learning enthusiasts to join us for this exciting challenge in a relaxed, collaborative environment. Participants will enhance their skills in data analysis, machine learning, and meteorological modeling. No prior experience with weather data is required.  Additionally, you will have the opportunity to connect with like-minded individuals and work with KIT´s supercomputer cluster. 
This event is organized by machine learning enthusiasts from KCDS, with support from MathSEE, TRIANGEL and SCC.

Accurately estimating rainfall by radar data is challenging because radars measure reflectivity rather than direct rainfall, and environmental variations further complicate this conversion. The RainQuest hackathon aims to address this problem by developing models that integrate precise point measurements from rain gauge data with radar reflectivity, which offers better measurements resolution. By combining these data sources, we aim to enhance the precision of precipitation estimates.
We invite all data science and machine learning enthusiasts to join us for this exciting challenge in a relaxed, collaborative environment. Participants will enhance their skills in data analysis, machine learning, and meteorological modeling. No prior experience with weather data is required.  Additionally, you will have the opportunity to connect with like-minded individuals and work with KIT´s supercomputer cluster. 
This event is organized by machine learning enthusiasts from KCDS, with support from MathSEE, TRIANGEL and SCC.

The event will be held in English and run for two days, on October 21 and 22, 2024.
 
Researchers are responsible for ensuring that their own conduct complies with the standards of good research practice. The workshop will introduce basic issues of research integrity by addressing important guidelines of the Deutsche Forschungsgemeinschaft (DFG) and specific regulations of KIT for safeguarding good research practice – relevant for every early career researcher@KIT.
 
A) Basics of Responsible Conduct
Introduction: Research, ethical principles and professional ethos of a researcher Basic (inter-)national recommendations and regulations for safeguarding good research practice Research misconduct: Examples, elements of offense, reasons and consequences  
B) General Responsibilities
Quality management: research design, documentation/archiving Publication process, authorship and review of manuscripts Supervision: Expectations/duties/roles Organizational culture: Collaboration, communication, prevention and dealing with conflict Procedures in case of suspicion and relevant contact points  
C) Important Specific Responsibilities
Important prior to any data collection: Authorization or permission relevant research Possible topics (depending on participants’ disciplinary/research background): ‘Research on animals’, ‘Research on humans’ and/or ‘Surveys, interviews, data privacy and security issues in research’  
Via dialogic inputs, discussion of case examples, single/group work and plenary discussion participants will have the opportunity to discuss and reflect their individual research practice and professional attitudes on being a researcher.
 
This event is open to doctoral researchers and postdocs at KIT who are KHYS members.
 
The event will be held in English and run for two days, on October 21 and 22, 2024.
 
Technical requirements: To participate in this event, you need a stable internet connection, a webcam and a microphone. Participants will receive further detailed information regarding the online-platform prior to the event.

If you are unable to attend an event, please inform us promptly via e-mail. This way you are allowing your colleagues the opportunity to participate and you help us to maintain the quality of our Further Education Program. Thank you!

This two-day on-site workshop takes place on November 21 & 22, 2024.
 
Content
“Do I have a plan for how to achieve my PhD? How do I define priorities and draw up a schedule? What is required for the process of writing?” A PhD project lasting a number of years will raise many questions requiring individual and tailor-made solutions.
 
This workshop will help you develop your individual strategy ensuring that your planning will both be effective and efficient in achieving your goal. You will not only learn more about self-management methods, but will also identify and explore external resources such as dealing with the expectations of others and receiving active support from your supervisors.
 
Registration
Please visit the KCDS intranet portal to register for this course.
For non-members of KCDS: 
In order to be able to book the course, a KCDS guest/non-member account is necessary.
If you don't have an account yet, you can register here: Create a KCDS account to book courses
Once your account is activated by a KCDS administrator, you will be able to book the course.
 
Picture: Mindspace Studio on Unsplash

This two-day on-site workshop takes place on November 21 & 22, 2024.
 
Content
“Do I have a plan for how to achieve my PhD? How do I define priorities and draw up a schedule? What is required for the process of writing?” A PhD project lasting a number of years will raise many questions requiring individual and tailor-made solutions.
 
This workshop will help you develop your individual strategy ensuring that your planning will both be effective and efficient in achieving your goal. You will not only learn more about self-management methods, but will also identify and explore external resources such as dealing with the expectations of others and receiving active support from your supervisors.
 
Registration
Please visit the KCDS intranet portal to register for this course.
For non-members of KCDS: 
In order to be able to book the course, a KCDS guest/non-member account is necessary.
If you don't have an account yet, you can register here: Create a KCDS account to book courses
Once your account is activated by a KCDS administrator, you will be able to book the course.
 
Picture: Mindspace Studio on Unsplash

This two-day on-site workshop takes place on Januar 13 & 14, 2025.
 
Content
Publications of research results is the currency in modern science. It might not be your favourite occupation, but it is a decisive one, strongly determining your future research and career opportunities. In this course, you will be introduced to the process of writing a scientific paper. In order to deepen and apply this knowledge, you will also actually produce a manuscript based on your research data (and one that might serve as the foundation for a journal submission). Questions about other forms of text like conference papers can also be discussed.
 
A-Z of scientific writing
Developing a clear main message Priority setting in the writing process Structure and logical flow of the text Create meaningful graphics Memorable introduction and abstract How to develop a daily writing routine Efficient communication with co-authors Plus: Writing with AI tools
Please note: Given that this is a hands-on online-workshop, you need some (analysed) data that can serve as the foundation for a journal article. Please also bring some scientific texts written by you (e.g. Bachelor or Master thesis, as well as any published or upcoming papers) to the workshop.
 
Registration
Please visit the KCDS intranet portal to register for this course.
For non-members of KCDS: 
In order to be able to book the course, a KCDS guest/non-member account is necessary.
If you don't have an account yet, you can register here: Create a KCDS account to book courses
Once your account is activated by a KCDS administrator, you will be able to book the course.
 
Picture by Joanna Kosinska on Unsplash

This two-day on-site workshop takes place on Januar 13 & 14, 2025.
 
Content
Publications of research results is the currency in modern science. It might not be your favourite occupation, but it is a decisive one, strongly determining your future research and career opportunities. In this course, you will be introduced to the process of writing a scientific paper. In order to deepen and apply this knowledge, you will also actually produce a manuscript based on your research data (and one that might serve as the foundation for a journal submission). Questions about other forms of text like conference papers can also be discussed.
 
A-Z of scientific writing
Developing a clear main message Priority setting in the writing process Structure and logical flow of the text Create meaningful graphics Memorable introduction and abstract How to develop a daily writing routine Efficient communication with co-authors Plus: Writing with AI tools
Please note: Given that this is a hands-on online-workshop, you need some (analysed) data that can serve as the foundation for a journal article. Please also bring some scientific texts written by you (e.g. Bachelor or Master thesis, as well as any published or upcoming papers) to the workshop.
 
Registration
Please visit the KCDS intranet portal to register for this course.
For non-members of KCDS: 
In order to be able to book the course, a KCDS guest/non-member account is necessary.
If you don't have an account yet, you can register here: Create a KCDS account to book courses
Once your account is activated by a KCDS administrator, you will be able to book the course.
 
Picture by Joanna Kosinska on Unsplash