DISA Seminar December 6th: Machine Learning for Astroparticle Physics

Monday, November 22nd, 2021

Astroparticle physics is a sub-branch of Physics dealing with the detection of gamma-rays, neutrinos, gravitational waves and cosmic rays from the Universe. In this seminar, we will focus on the field of extragalactic gamma-ray astronomy.

The two main types of datasets in the field are those produced via simulations and those acquired via detector equipments (“real”), leading to a large amount of data to be calibrated, prepared, filtered, reconstructed and analysed. Simulations are especially needed in order to understand the sensitivity of the given equipment to a given searched physics “signal” and to prepare the data analysis procedure before looking at the real data. Decades of experience in data analysis in the field lead to the ability to publish solid results.

In this context, machine learning is giving a big boost in speeding up the data analysis procedures. The first successful applications of supervised Machine Learning in the field date back to the years 2010-2013 and concern classification and regression methods. Nowadays, there is a huge effort to exploit Deep Learning methods to achieve faster simulations and to improve the current data analysis methods, where physicists cannot “see” or “predict” any significant features describing the datasets.

I will present the current activities and future challenges of my research groups at Linnaeus University and at the University of Paris and, more generally, the challenges of the field.

Keywords: gamma-ray astronomy, supervised machine learning, event classification, feature regression


  1. Y. Becherini et al., 2011 (Astrop. Phys., Vol 34, 12, 2011, 858-870)
  2. Y. Becherini et al., 2012 (Gamma 2012)
  3. Y. Becherini et al., 2012 (arXiv:1211.5997) 
  4. M. Senniappan et al
  5. T. Bylund et al., 

Astroparticle physics and Big Data within DISA

Friday, June 2nd, 2017

One of the seven areas within DISA is Astroparticle physics and the research group are mainly focusing on the project ALTO at the moment. ALTO is a project the research group is developing for a wide-field Very High Energy (VHE) gamma-ray detector to be installed at very high altitude in the Southern hemisphere. The main work within the group is focused on:

  • The design of the detection tanks and of the full array.
  • The construction of a prototype detection unit on the Växjö campus, including choice of the photo-multipliers, of the scintillator and Cherenkov units optimization, and of the electronics needed to digitize the photo-multiplier signals.
  • The optimization of the array trigger.
  • The preparation of storage, transport and online and offline analyses of the “Big Data” generated by the experiment which will take data 24h per day with no interruptions.

The ALTO research team at the spot where a prototype one detector element will be built. From left to right: Satyendra Thoudam, Yvonne Becherini, Jean-Pierre Ernenwein, Michael Punch


During the period May- September 2017 the group has an invited Guest Professor from University of Marseille, Jean-Pierre Ernenwein.

For more information on ALTO see the university website or the project website. If you have any questions about the project, please contact the project leader Yvonne Becherini.


5 new PhD students wanted!

Monday, May 15th, 2017

Linnaeus University Center for Data Intensive Science and Applications (DISA) has taken the first steps to build new competencies to the newly established research environment by opening up five positions for PhD students.

  • Physics with specialization in “Big Data in Astroparticle Physics”
  • Computer Science with specialization in Software and Information Quality
  • Computer Science with specialization in Visual Analytics
  • Computer Science with specialization on eHealth
  • Computer Science with specialization on formal methods and Security