Machine Learning Tools for Research in Astronomy

Harnessing the tools of Machine Learning methods for analysis and discovery in observations and simulations

Scientific Organizing Committee

Annalisa Pillepich (MPIA)
Gianfranco Bertone (U. of Amsterdam)
David W Hogg (NYU, Flatiron Inst.)
Kai Polsterer (HITS)
Michelle Ntampaka (Harvard)
Glenn van de Ven (U. of Vienna)

Local Organizing Committee

Elad Zinger
(+49|0) 6221 528-221

Email the LOC



Camille Avestruz (U. of Michigan)
Dalya Baron (TAU)
Vincent Boucher (B12 Consulting)
Massimo Brescia (INAF/Napoli)
Tobias Buck (AIP)
Tom Charnock (IAP)
Gabriella Contardo (CCA)
Erin Cram (Northeastern U.)
Joachim Denzler (Jena U.)
Lukas Eisert (MPIA)
Jesus Falcon Barroso (IAC)
Doug Finkbeiner (Harvard)
Morgan Fouesneau (MPIA)
Neige Frankel (MPIA)
Timothy Gebhard (MPI Intelligente Systeme)
Nikos Gianniotis (HITS, Heidelberg)
Andy Goulding (Princeton)
Gregory Green (MPIA)
David W. Hogg (NYU; CCA)
Marc Huertas-Company (Paris Obs./U. Paris Diderot.)
Prashin Jethwa (U. of Vienna)
Nikolay Kacharov (MPIA)
Tomasz Kacprzak (ETH Zurich)
Wolfgang Kerzendorf (NYU/Michigan State)
Mikhail Kovalev (MPIA)
Victor Ksoll (U. of Heidelberg)
Francois Lanusse (UC Berkely)
Luisa Lucie-Smith (UCL)
Peter Melchior (Princeton)
Brice Menard (Johns Hopkins U.)
Allison Merritt (MPIA)
Brian Nord (Fermilab)
Michelle Ntampaka (Harvard)
Stella Offner (U. of Texas)
Mario Pasquato (INAF Padova)
Josh Peek (Space Telescope Institute)
Laurence Perreault Levasseur (U. of Montreal)
Annalisa Pillepich (MPIA)
Kai Polsterer (HITS, Heidelberg)
Stephen Portillo (U. of Washington)
Dov Poznanski (TAU)
Nesar Ramachandra (Argonne Labs)
Nima Sedaghat (ESO)
Glenn van de Ven (U. of Vienna)
Soledad Villar (NYU; CDS)
Ashley Villar (Harvard)
Yuan-Sen Ting (IAS; ANU)
Elad Zinger (MPIA)


Machine learning methods can reveal the relations between observational data and the properties of the observed objects. They simultaneously have the potential to complement and inform the build up of physical models, provide data-driven predictions for practical applications, and automatize the analysis and comparisons of large data sets (real or simulated).

A number of machine learning tools have already been proven successful in a variety of astrophysical contexts, e.g. from the measurement of galaxy morphologies to automatic object detection and identification (e.g. quasars in the early Universe or RR Lyrae stars around the Milky Way), streamlined estimation of galaxies’ redshifts, and image analysis techniques for de-noising and deconvolution. In the upcoming years efficient machine learning and image analysis algorithms will become of the essences to fully exploit the available data and make substantial progress in our understanding of astrophysical phenomena.

We are therefore planning a hands-on and interactive workshop to foster big-data and machine-learning proficiency in the astronomy community, with particular focus on applications, among others, pertaining extra-galactic and galactic astronomy, galaxy formation, large scale structure, cosmology and cosmological surveys. Tools will range from classical regression and classification methods to state-of-the-art deep learning algorithms as generative adversarial networks and variational autoencoders. We plan to invite to participate also a few representatives of big and small private companies, in order to favor the percolation of their know-how expertise to the astronomy community.

The meeting will focus on the following subjects:

  • Obtaining an overview of the current landscape of ML applications in astronomy
  • Facilitating an interface between data science-oriented astronomers, astrophysicists and cosmologists with every-day practitioners and researchers in machine learning techniques.
  • Enhancing the sophistication and rigor of the analyses and discovery potential of astronomical data, both from observations and computer simulations.
  • Introducing input and inspirations from applications of ML in other fields

    Talks & Slides

    Inrtroduction & Opening Statements
    Participant Introduction
    Dalya Baron: Finding simple structures in complex dataset
    Vincent Boucher: Applications of machine learning in industry, a consulting company perspective (missing)
    Massimo Brescia: A sustainable synergy between Astrophysics and Data Science
    Tobias Buck: PICASSO: Painting IntrinsiC Attributes onto SDSS Objects
    Tom Charnock: Modern machine learning methods for trustworthy science
    Gabriella Contardo: Representation learning and cosmological data (missing)
    Joachim Denzler: Active and Life-Long Learning: Elements of Future Machine Learning Systems
    Jesus Falcon Barroso: Deciphering the hidden side of galaxies with full-spectral fitting
    Doug Finkbeiner: GSPICE: Artifact Detection and Repair in Spectral Data using Gaussian Processes (missing)
    Morgan Fouesneau: ML in Gaia Data processing
    Timothy Gebhard: Half-Sibling Regression meets Direct Imaging: A Causal Approach for Uncovering Exoplanets (missing)
    Nikos Gianniotis: Mixed Variational Inference
    Andy Goulding: Identifying the interesting needles in astronomical haystacks (missing)
    Gregory Green: Data-driven stellar spectral energy distributions
    David W. Hogg: The causal structure of ML methods in astronomy
    Marc Huertas-Company: Comparison of simulations of galaxy formation and observations with deep learning
    Prashin Jethwa: Restoring latently-lost meaning in population-dynamical galaxy decompositions
    Nikolay Kacharov: Chemical abundances from low resolution MUSE spectra in the Sgr nucleus
    Tomasz Kacprzak: Cosmological analysis of weak lensing maps with deep learning
    Wolfgang Kerzendorf: Emulators - a link between physics and machine learning
    Mikhail Kovalev: Machine learning in spectroscopy
    Victor Ksoll: Characterising Pre-Main-Sequence Stars in the Large Magellanic Cloud with ML and Deep Learning
    Francois Lanusse: Hybrid Physical-Deep Learning Modeling from Large Scale Structure to Galaxy Morphology
    Luisa Lucie-Smith: Machine learning the formation of dark matter halos
    Peter Melchior: Solving inverse problems with hard constraints and data-driven priors
    Brice Menard: Feature_visualization
    Brice Menard: Describing Complexity
    Brian Nord: Disrupted by Data Science: A turning point for Science and Society (missing)
    Michelle Ntampaka: Barriers and Opportunities for ML in Astronomy
    Michelle Ntampaka: A Hybrid Deep Learning Approach to Cosmological Constraints From Galaxy Redshift Surveys
    Stella Offner: Harnessing Machine Learning to Study the Life Cycle of Stars
    Mario Pasquato: Training on simulated data: a few case studies
    Josh Peek: Approaches to Deep Learning-Assisted Discovery
    Laurence Perreault Levasseur: Analysis of Gravitational Lensing Data with Machine Learning (missing)
    Kai Polsterer: From Photometric Redshifts to Improved Weather Forecasts: an interdisciplinary view on astroinformatics (missing)
    Stephen Portillo: Dimensionality Reduction of SDSS Spectra with Autoencoders
    Dov Poznanski: Using unsupervised ML for discovery
    Nesar Ramachandra: Bridging the reality gap in Machine Learning with synthetic data training
    Soledad Villar: Adversarial examples in machine learning models of stellar spectra
    Ashley Villar: Time Domain Astrophysics in the Era of LSST
    Yuan-Sen Ting: Machine learning meets statistics: Lessons from image emulation(missing)

    Workshop Venue

    Schloss Ringberg

    The workshop will be held in the Schloss Ringberg, overlooking the Tegernsee Lake, in Southern Bavaria. Lodging and meals will be provided at the castle as well. For more information on the castle see here.

    Meals and Lodging

    All participants will be lodged at the Castle for the duration of the program.

    Breakfast, Lunch, and Dinner will all be provided at the Castle as well. The first meal will be a light dinner on Sunday evening. The last meal served will be Lunch on Friday, just before departure. Vegetarian options will be possible as well.


    Code of Conduct

    The SOC and LOC are committed to creating an environment that is safe, professional and of mutual trust where diversity and inclusion are valued, and where everyone is entitled to be treated with courtesy and respect. As conference organisers we are committed to making this workshop, and all associated activities productive and enjoyable for everyone. We will not tolerate harassment of participants in any form.

    Please follow these guidelines:

  • Behave professionally. Harassment and sexist, racist, or exclusionary comments or jokes are not appropriate. Harassment includes sustained disruption of talks or other events, inappropriate physical contact, sexual attention or innuendo, deliberate intimidation, stalking, and photography or recording of an individual without consent. It also includes offensive comments related to individual characteristics, for example: age, gender, sexual orientation, disability, physical appearance, race, nationality or religion.
  • All communication should be appropriate for a professional audience including people of many different backgrounds. Sexual or sexist language and imagery is not appropriate.
  • Be respectful and do not insult or put down other attendees or facilitators of the event. Critique ideas not people.
  • Should a participant witness events of bullying, harassment or aggression, we recommend that they approach the affected person to show support and check how they are. The witness may also wish to suggest that the person report the inappropriate behaviour. However, it is up to the affected person alone whether or not they wish to report it.
  • If participants wish to share photos of a speaker on social media, we strongly recommend that they first get the speaker’s permission. Participants may also share the contents of talks/slides via social media unless speakers have asked that specific details/slides not be shared.
  • The designated contact points for any issue relating to this Code of Conduct, from the SOC and LOC are Annalisa Pillepich and Elad Zinger. Participants can report any violation of these guidelines to these designates in confidence. If asked to stop inappropriate behaviour participants are expected to comply immediately and, in serious cases, may be asked to leave the event. The organizers will not tolerate retaliation against anyone reporting violations of this code of conduct.

    Acknowledgement: This code of conduct is based on the ESO Code of Conduct for Workshops & Conferences.