Regensburg 2019 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

SOE: Fachverband Physik sozio-ökonomischer Systeme

SOE 4: Computational Social Science and Data Science II

SOE 4.3: Talk

Monday, April 1, 2019, 12:30–12:45, H17

Big Data and Machine Learning in Astrophysics — •Karl Mannheim — Universität Würzburg, Lehrstuhl für Astronomie

Next-generation observatories such as SKA will produce data at a rate higher than can be analyzed by human scientists. Therefore, fast analysis methods such as machine learning will play a major role in the near future. Methods applied to offline data target the extraction of scientifically relevant data from the raw data, e.g. by classifying objects or recognizing morphological patterns in images or in the spectral domain. Methods applied to online data can feedback with the data aquisition system to optimize the detector performance or reduce the data volume by filtering out unusable low-quality data. Utilizing their full potential requires to advance machine learning algorithms from finding correlations to allowing causal inferences. Users and developers of ML methods from all branches of physics, astronomy, or computers science are encouraged to discuss and exchange ideas in the new DPG working group AKPIK.