[LMU-OSC News] Hackathon at ifo 28.-30. April

LMU Open Science Center News lmu-osc at lists.lrz.de
Thu Feb 23 09:11:05 CET 2023


Dear OSC members,

The Information and Forschung Institute for Economic Research (ifo) is organizing a 2-day hackathon on 28.-30. April - the ifoHack 2023. You can find more information on the event page https://www.ifo.de/en/ifohack2023.

Registration and feedback are required before 3rd March 2023.

Please forward the information below to your department as the event is open for everyone (Seniors/Post-Docs, Juniors/PhD students, specialists, interns, research assistants).

What is a hackathon?
Teams from different backgrounds ranging from economics and statistics, psychology, sociology, computer science, design, and engineering will develop prototypical solutions to economic and social problems using state-of-the-art technology, tools and infrastructure.

Why ifoHack?
There are large potential gains from combining sharp economic reasoning and pre-existing technical knowledge with basic programming in Python and modern web technologies. The way how economic research is conducted has changed remarkably and new hypotheses require an appropriate, modern approach to empirical and analytical applications. Replicability and transparency are increasingly important to society but also within academia, especially with respect to the publication process.

The ifoHack is an ideal environment for testing first ideas that could lead to innovative projects and publications, to explore or to construct a novel dataset, to build prototypes that could automate tasks, to develop a predictive model that performs better than conventional models at relatively low costs. Lastly, it can be an enjoyable experience to collaborate with colleagues you didn't work together with in the past, to learn from each other and to hold a "tangible" product (such as a dashboard) in your hands that you can show to the world after a very short period! This event is meant to provide a low-barrier opportunity for exposing yourself to unknown concepts in a safe (but nonetheless intense) environment. Only basic knowledge and experience of programming with Python (~1 year) is a plus.

Which kind of challenges to expect?
Participants may come up with their own ideas or select a pre-defined but open-ended challenge each which is mentored by one of the ifo departments. For example, teams develop

  *   a predictive model for firms' answers from the ifo Business Survey based on micro data.
  *   a dashboard of real-time indicators that can be used to nowcast economic variables, deploy it and provide a user interface to obtain predictions given inputs.
  *   a method to extract information from texts capturing sentiment, inter-linkages and/or factors for predicting stock price changes.
  *   a language model and interface ("ifoGPT", similar to ChatGPT) trained on press releases for text generation.
  *   an interactive map of households' characteristics at very high resolution, involving spatial computations and satellite imagery.

If you (or your department) are (is) considering a project idea, a research question or a specific task to be solved in a new challenge that requires certain computational resources or specific knowledge - please send a very short description (2-3 sentences) to ifohack at ifo.de<mailto:ifohack at ifo.de> or discuss your proposal directly with the organisor Gerome Wolf (wolf.g at ifo.de<mailto:wolf.g at ifo.de>) until 3rd March 2023.

Why participate?
Attending a hackathon is an enriching and sustainable experience in and of itself and highlights your ability to implement a technical solution within a short amount of time and in a collaborative manner which will also improve your CV. You will acquire new skills that lend themselves to ambitious research projects and are in high demand for data-centric industry roles such as orchestration of data pipelines, reviewing code, and deploying your own application.
There are also great prizes for the winning teams to assist you in following -up on the concepts you encounter during the hackathon.

You will benefit greatly and should attend if you want to...

  *   have an unforgettable experience, a steep learning curve, and lots of fun!
  *   adopt an algorithmic way of thinking.
  *   try out an ubiquitous but unfamiliar concept (e.g. GPT) on real data.
  *   like-minded hackers and work on meaningful challenges.
  *   brush up or upgrade your operational research skills by integrating modern open-source tools.
  *   leverage computational capacity to complete tasks more efficiently.

Food, hacking spaces, exciting data, and infrastructure will be provided - expect nice surprises!

When and where does it take place?
The ifoHack 2023 will take place from Friday 28th (3pm) - Sunday 30th April 2023 (5pm) and will be hosted in-person at the ifo premises.

A registration for the hackathon is required under https://ifohack2023.eventbrite.de as seats are limited to 30 (first come, first served).

We're looking forward to ifoHack 2023 and your ample participation!
In case of questions, please contact ifohack at ifo.de<mailto:ifohack at ifo.de> or Gerome Wolf (wolf.g at ifo.de<mailto:wolf.g at ifo.de>) directly.

Kind regards,
The ifoHack 2023 team

____________
ifo Institut - Leibniz-Institut fuer Wirtschaftsforschung an der Universitaet Muenchen e.V.
Poschingerstr. 5, 81679 Muenchen,
Sitz: Muenchen, Vereinsregister-Nr.: 4419, Amtsgericht Muenchen,
Vorstand: Prof. Dr. Dr. h.c. Clemens Fuest (Praesident), Dr. Stephanie Dittmer;
Steuernummer 143/217/10159, USt-IdNr. DE129516729

Folgen Sie dem ifo Institut auf Twitter: https://twitter.com/ifo_Institut
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lrz.de/pipermail/lmu-osc/attachments/20230223/425ddbaa/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Plakat_ifoHack.pdf
Type: application/pdf
Size: 153723 bytes
Desc: Plakat_ifoHack.pdf
URL: <http://lists.lrz.de/pipermail/lmu-osc/attachments/20230223/425ddbaa/attachment.pdf>


More information about the lmu-osc mailing list