AG Splinter Meeting 2025:
E-Science & E-Infrastructures & Machine Learning & Virtual Observatory

Annual Meeting 2025 of the Astronomische Gesellschaft, Berlin, Germany

E-Science & E-Infrastructures & Machine Learning & Virtual Observatory

This splinter meeting is dedicated to standard infrastructures for astronomical data dissemination and analysis, with an extra focus on Machine Learning and other AI techniques. These are, on the one hand, particularly data-hungry and hence depend on the availability of easily processable, well-described data at scale. On the other hand, generative techniques (not at all restricted to LLMs) promise to address some limitations of our current systems, from flattening admittedly steep learning curves to contributing to more focused data discovery.

A focus topic this time will be resources for training astronomers to efficiently use the data and software infrastructures built over the past decades. Speakers presenting on courseware, novel ways of knowledge dissemination or just sharing experiences made in teaching data science will be most welcome. Another obvious topic will address progress on making astronomical data even FAIR-er than it already is. There is the NFDI in Germany with PUNCH, ErUM Data, the formation of the DZA, and there are more astro-infrastructure projects and efforts going on. This splinter meeting is dedicated to standard infrastructures for data dissemination and analysis, with an extra focus on Machine Learning as a particular data-hungry field with high relevance to essentially all areas of astronomy.

We welcome contributions on applying existing and emerging technologies as well as reports from the frontiers of federating information systems to facilitate astronomical research.

Tuesday,   16.09.2025, 14:00 - 16:30, (Schlesische Museum)

Convenors

Markus Demleitner (ARI), K. Polsterer (H-ITS), M. Hoeft (TLS), H. Enke (AIP)

Agenda and Presentations

Tuesday,   16.09.2025


Bernhard Schulz (SOFIA Data Center, IRS, Universität Stuttgart) Progress in Constructing the SOFIA Data Center
Marco Bischoff (Forschungszentrum Jülich) Quantifying Model-Observation Similarity: Implications for Stellar Flyby Models
Markus Demleitner (ARI, U. Heidelberg) Teaching the Virtual Observatory
E. Tom L. Strauß (AIP) Compute Cloud: Constructing a high available container orchestration infrastructure
Iliana Cortes (HITS) Photometric decomposition of AGN variability via Machine Learning: Spectral reconstruction and lag characterization
Johanna Riedel (HITS) Back to the feature: Reconstructing Spectra from Photometry
Sebastian T. Gomez (HITS) From data to scientific breakthroughs with tools powered by Representation Learning
Romain Chazotte (HITS) Improved Rotation Equivariance via Embedding in Zernike Polynomial Space

Contact

If you have any further questions, please don't hesitate to contact us

Kai Polsterer: kai.polsterer [at] h-its [dot] org
Markus Demleitner: msdemlei [at] ari [dot] uni-heidelberg [dot] de
Matthias Hoeft: hoeft [at] tls-tautenburg [dot] de
Harry Enke: henke [at] aip [dot] de

EScience & Virtual Observatory Splinters at AG Meetings (2013-2025)