Debating the potential of machine learning in astronomical surveys

Planning

Week Mon. 18 Tue. 19 Wed. 20 Thu. 21 Fri. 22 List

Monday, October 18, 2021

Time	Event	(+)
08:45 - 09:45	In person Registration - Registration of in person participants
09:45 - 10:00	Welcome speech (Henri Mineur amphitheater / Plenary online session) - Organizers
10:00 - 11:00	Review (Henri Mineur amphitheater / Plenary online session)	(+)
10:00 - 11:00	› Review on cosmology and machine learning - Benjamin Wandelt, Institut d'Astrophysique de Paris
11:00 - 11:20	Break (Breakout rooms)
11:20 - 12:20	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
11:20 - 11:35	› Probabilistic Mapping of Dark Matter by Neural Score Matching - Benjamin Remy, Cosmostat, AIM, IRFU, CEA-Saclay
11:35 - 11:50	› Simulation and Segmentation with Deep Learning for Euclid - Hubert Bretonnière, Institut d'astrophysique spatiale
11:50 - 12:05	› Search for galaxy-scale strong lenses in DES and CFIS - Frédéric Courbin, EPFL, Laboratoire d'astrophysique, Ecole Polytechnique Fed ´ erale de Lausanne
12:05 - 12:20	› Learning from 3D tomographic 21cm maps - Caroline Heneka, Hamburg Observatory, Hamburg University
15:00 - 15:20	Flash talk (Henri Mineur amphitheater / Plenary online session)	(+)
15:00 - 15:04	› Unifying large-scale spectroscopy, astrometry, and photometry with Convolutional Neural-Networks. - Guillaume Guiglion, Leibniz-Institut für Astrophysik Potsdam
15:04 - 15:08	› Unknown Unknowns: Hybrid machine learning and template based photometric redshifts - Peter Hatfield - University of Oxford
15:08 - 15:12	› A Deep Learning Neural Network for Voigt Profile Fitting Quasar Absorption Lines - Bryson Stemock - New Mexico State University
15:12 - 15:16	› Accelerating the modeling of HI on cosmological scales via Deep Learning - Robert Feldmann - University of Zürich [Zürich]
15:16 - 15:20	› Beyond the Hubble Sequence - Exploring galaxy morphology with unsupervised machine learning - Ting-Yun Cheng, Durham University
15:20 - 16:05	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
15:20 - 15:35	› Fast and realistic large-scale structure from machine-learning-augmented random field simulations - Davide Piras, UCL
15:30 - 16:05	› Parameter Estimation with Physics Informed Neural Networks - Alex Lague, University of Toronto, Canadian Institute for Theoretical Astrophysics (CITA)
15:35 - 15:50	› A new catalog of 343,000 quasars with their photometric redshifts derived with machine learning from the Kilo Degree Survey - Szymon Nakoneczny, National Centre for Nuclear Research [Otwock]
16:05 - 16:25	Break (Breakout rooms)
16:25 - 17:25	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
16:25 - 16:40	› Machine learning-infused cluster cosmology - Stéphane Ilic, AstroParticule et Cosmologie, Laboratoire dÉtude du Rayonnement et de la Matière en Astrophysique, Institut de recherche en astrophysique et planétologie
16:40 - 16:55	› CosmoPower: deep learning emulation of cosmological power spectra for accelerated Bayesian inference from next-generation surveys - Alessio Spurio Mancini, University College, London
16:55 - 17:10	› Single frequency CMB polarized foreground marginalization with a single training image - Niall Jeffrey, École normale supérieure
17:10 - 17:25	› Bayes vs. machine learning in large-scale survey demographics with eROSITA - Johannes Buchner, Max Planck Institute for Extraterrestrial Physics
17:25 - 18:00	Break (Breakout rooms)
18:00 - 19:30	Debate (Henri Mineur amphitheater / Plenary online session)
19:30 - 22:30	Cocktail dinner - Cocktail of the conference in Cassini room for in person participants - Cocktail dinner

Tuesday, October 19, 2021

Time	Event	(+)
10:00 - 11:00	Review (Henri Mineur amphitheater / Plenary online session)	(+)
10:00 - 11:00	› Imaging Surveys and ML - Elisabeth Krause, University of Arizona
11:00 - 11:20	Break (Breakout rooms)
11:20 - 12:20	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
11:20 - 11:35	› Source detection through semantic segmentation with convolutional neural networks - Maxime Paillassa, Laboratoire dÁstrophysique de Bordeaux [Pessac], Division of Physics and Astrophysical Science, Graduate School of Science, Nagoya University
11:35 - 11:50	› Identifying strong gravitational lenses in current and future large-scale imaging surveys - Raoul Canameras, MPA, Garching
11:50 - 12:05	› Reconstructing blended galaxies with Machine Learning - Lavanya Nemani, INAF - Osservatorio Astronomico di Roma
12:05 - 12:20	› Pushing automated morphological classifications to their limits with the Dark Energy Survey - Jesús Vega Ferrero, Instituto de Astrofisica de Canarias
15:00 - 15:20	Flash talk (Henri Mineur amphitheater / Plenary online session)	(+)
15:00 - 15:04	› Rediscovering Newton's gravity and Solar System properties using deep learning and inductive biases - Pablo Lemos, Department of Physics and Astronomy [UCL London], Department of physics and Astronomy [University of Sussex]
15:04 - 15:08	› Gaussian Process Regression: An Application to Radio Cosmology - Paula Soares, Queen Mary, University of London
15:08 - 15:12	› Detection of the Damped Lyman-alpha systems in quasar spectra with machine learning algorithms - ting tan, Laboratoire de Physique Nucléaire et de Hautes Énergies
15:12 - 15:16	› Forecasting the power of Higher Order Weak Lensing Statistics with automatically differentiable simulations - Denise lanzieri, CEA-Saclay
15:16 - 15:20	› Manifold learning to explore the galaxy parameter space - Iary Davidzon, DAWN/NBI - Keerthana Jegatheesan, Universidad Diego Portales - Olivier Ilbert, LAM - Clotilde Laigle, IAP
15:20 - 16:05	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
15:20 - 15:35	› Machine learning within the THREEHUNDERD simulation project - Daniel de Andres, Universidad Autonoma de Madrid
15:35 - 15:50	› Superresolving Herschel imaging: a proof of concept using Deep Neural Networks - Lynge Lauritsen, The Open University
15:50 - 16:05	› Capturing the physics of MaNGA galaxies with self-supervised Machine Learning - Regina Sarmiento, Universidad de La Laguna [Tenerife - SP], Instituto de Astrofisica de Canarias
16:05 - 16:25	Break (Breakout rooms)
16:25 - 17:40	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
16:25 - 16:40	› ‘Deep' vs. 'Shallow' Learning in Cosmological Surveys - Ofer Lahav, University College London
16:40 - 16:55	› A Method to Distinguish Quiescent and Dusty Star-forming Galaxies with t-SNE - John Weaver, Cosmic Dawn Center
16:55 - 17:10	› From machine learning to human understanding - Daniel Masters, Caltech / IPAC
17:10 - 17:25	› Use data not models -- Lensing of '69 as an example for data-driven inference - Jenny Wagner, https://thegravitygrinch.blogspot.com
17:25 - 17:40	› Machine learning: lessons learnt with the QUBRICS survey - Francesco Guarneri, Astronomical Observatory of Trieste, Dipartimento di Fisica [Trieste]
17:00 - 18:00	Break (Breakout rooms)
18:00 - 19:30	Debate (Henri Mineur amphitheater / Plenary online session)

Wednesday, October 20, 2021

Time	Event	(+)
15:00 - 16:00	Review (Henri Mineur amphitheater / Plenary online session)	(+)
15:00 - 16:00	› Review on generative models in Machine Learning - Gilles Loupe, Universite de Liege
16:00 - 16:20	Break (Breakout rooms)
16:20 - 16:40	Flash talk (Henri Mineur amphitheater / Plenary online session)	(+)
16:20 - 16:24	› Studying Morphology & Quenching of Galaxies in the All Sky-Era using Interpretable Bayesian Convolutional Neural Networks - Aritra Ghosh, Yale University
16:28 - 16:32	› Emulating the void density profile and the void-size function - Giorgia Pollina, Universitäts-Sternwarte, Ludwig-Maximilians Universität München
16:32 - 16:36	› Pix2Prof: fast extraction of sequential information from galaxy imagery via a deep natural language 'captioning' model - Michael Smith - Queen's University, Kingston, ON, Canada, University of Hertfordshire, Hatfield, UK
16:36 - 16:40	› Bayesian Inference Acceleration: a Deep Learning Approach - Marco Bonici, Istituto di Astrofisica Spaziale e Fisica Cosmica - Milano
16:40 - 17:40	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
16:40 - 16:55	› CosmicRIM : Reconstructing Early Universe by Combining Differentiable Simulations with Recurrent Inference Machines - Chirag Modi, Flatiron Institute
16:55 - 17:10	› Bridging the gap between simulations and survey data - domain adaptation for deep learning in astronomy - Aleksandra Ciprijanovic, Fermi National Accelerator Laboratory
17:10 - 17:25	› Stratified Learning: A general-purpose method for learning under covariate shift with applications to observational cosmology - Roberto Trotta, Department of Physics [Imperial College London], Scuola Internazionale Superiore di Studi Avanzati / International School for Advanced Studies
17:25 - 17:40	› Neural spectrum encoding - Peter Melchior, Princeton University
17:40 - 18:00	Coffee break (Breakout rooms)
18:00 - 19:30	Debate (Henri Mineur amphitheater / Plenary online session)

Thursday, October 21, 2021

Time	Event	(+)
10:00 - 11:00	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
10:00 - 10:15	› The Dark Quest project for cosmological emulation - Takahiro Nishimichi, Kavli Institute for the Physics and Mathematics of the Universe [Tokyo], Yukawa Institute for Theoretical Physics
10:15 - 10:30	› The scattering transform in cosmology, or, a CNN without Training - Sihao Cheng, Johns Hopkins University
10:30 - 10:45	› Considerations for optimizing photometric classification of supernovae from the Rubin Observatory - Catarina S. Alves, University College, London
10:45 - 11:00	› Machine Learning Calibration of Cosmic Shear Redshift Distributions - Angus Wright, Ruhr University Bochum
11:00 - 11:20	Interaction session (Breakout rooms)
11:20 - 12:05	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
11:20 - 11:35	› LyAl-Net: A high-efficiency Lyman-α forest simulation with neural network - Chotipan Boonkongkird, Sorbonne Universite
11:35 - 11:50	› Dark Substructure Sensitivity in the Euclid Survey with Machine Learning - Conor O'Riordan, Max Planck Institute for astrophysics
11:50 - 12:05	› Inferring the assembly and merger histories of galaxies with the IllustrisTNG simulations and machine learning - Lukas Eisert, Max Planck Institute for Astronomy
12:05 - 12:25	Flash talk (Henri Mineur amphitheater / Plenary online session)	(+)
12:05 - 12:09	› Applying Deep Neural Network to dark matter halo catalogues to constrain the dark energy equation-of-state parameters - Farida Farsian, INAF - Osservatorio Astronomico di Bologna, Alma Mater Studiorum University of Bologna
12:09 - 12:13	› Cosmological constraints from galaxy-galaxy lensing and galaxy clustering with HSC-Y1 and BOSS data: the first application of emulator-based halo model to cosmology analysis - Hironao Miyatake, Kavli Institute for the Physics and Mathematics of the Universe [Tokyo], Nagoya University
12:13 - 12:17	› Tidal stream detection in HSC-SSP with Deep Learning - Helena Domínguez Sánchez, Institute of Space Sciences [Barcelona]
12:17 - 12:21	› Using convolutional neural networks to identify strong lenses in Euclid and J-PAS - Alberto Manjon García, Instituto de Física de Cantabria
12:21 - 12:25	› Galaxy Cluster Mass Estimation Using Deep Learning - Matthew Ho, Carnegie Mellon University
15:00 - 16:00	Review (Henri Mineur amphitheater / Plenary online session)	(+)
15:00 - 16:00	› Review on machine learning and cosmological simulations - Francisco Villaescusa-Navarro, Flatiron Institute, Department of Astrophysical Sciences [Princeton]
16:00 - 16:20	Coffee break (Breakout rooms)
16:20 - 16:40	Flash talk (Henri Mineur amphitheater / Plenary online session)	(+)
16:20 - 16:24	› Photometric redshift calibration with simulated annealing and self-organising maps - Benjamin Stölzner, University College, London
16:24 - 16:28	› Semi-Supervised Learning for Probabilistic Photometric Redshifts - Rory Coles, National Research Council of Canada, Herzberg Astronomy & Astrophysics Program, University of Victoria
16:28 - 16:32	› Time Delay Estimation in Unresolved Lensed Quasars via Convolutioanl Neural Networks - Luca Biggio, ETH - Alba Domi, University of Genoa - Luca Paganin, University of Genoa
16:32 - 16:36	› Detection of the Damped Lyman-alpha systems in quasar spectra with machine learning algorithms - ting tan, Laboratoire de Physique Nucléaire et de Hautes Énergies
16:35 - 17:20	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
16:35 - 16:50	› Real-time science-driven follow-up of survey transients: Optimally augmenting ZTF SN Ia light curves for maximizing SALT2 parameter constraints - Niharika Sravan - Caltech
16:50 - 17:05	› What drives the scatter in the BPT diagrams? A Machine Learning based analysis - Mirko Curti - Kavli Institute for Cosmology [Cambridge]
17:05 - 17:20	› Quantifying uncertainty in deep-learning classification of radio galaxies - Fiona Porter - Jodrell Bank Centre for Astrophysics
17:20 - 18:00	Coffee break (Breakout rooms)
18:00 - 19:30	Debate (Henri Mineur amphitheater / Plenary online session)
20:30 - 23:00	Conference Dinner - Conference dinner pour in person participation

Friday, October 22, 2021

Time	Event	(+)
10:00 - 11:00	Review (Henri Mineur amphitheater / Plenary online session)	(+)
10:00 - 11:00	› Review on Exoplanets and Machine Learning - Ingo Waldmann, University College London - London's Global University
11:00 - 11:20	Interaction session (Breakout rooms)
11:20 - 12:20	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
11:20 - 11:35	› An hybrid physics-ML framework to model exoplanetary light curves - Mario Morvan, University College London - London's Global University
11:35 - 11:50	› A deep transfer learning approach to photospheric parameters of CARMENES target stars - Antonio Bello-Garcia, University of Oviedo
11:50 - 12:05	› Multifrequency Point Source detection with Fully-Convolutional Networks: Performance in realistic microwave sky simulations - José Manuel Casas, Instituto Universitario de Ciencias y Tecnologías Espaciales de Asturias (ICTEA), University of Oviedo, Department of Physics
12:05 - 12:20	› Using a series Machine Learning models for the detection of high-redshift Radio Galaxy candidates. - Rodrigo Carvajal, Institute of Astrophysics and Space Sciences - University of Lisbon
15:00 - 16:00	Review (Henri Mineur amphitheater / Plenary online session)	(+)
15:00 - 16:00	› Review on Machine Learning for Radio-astronomy - Michelle Lochner, University of the Western Cape, South African Radio Astronomy Observatory, African Institute for Mathematical Sciences
16:00 - 16:20	Coffee break (Breakout rooms)
16:20 - 17:20	Contributed (Henri Mineur amphitheater / Plenary online session)	(+)
16:20 - 16:35	› hyphy: Mapping Dark Matter to Hydrodynamics with Posterior Inference - Benjamin A Horowitz, Department of Astrophysical Sciences [Princeton]
16:35 - 16:50	› MINERVA team: Winning the SKA Science Data Challenge 2 with fast dedicated CNN architectures - David Cornu, Laboratoire d'Etude du Rayonnement et de la Matière en Astrophysique
16:50 - 17:05	› Inference and probabilistic modelling with machine learning for LISA data analysis - Natalia Korsakova, Systèmes de Référence Temps Espace
17:05 - 17:20	› Deep Neural Networks for detection of sources in radio astronomical maps: main challenges and current achievements. - Renato Sortino, INAF - Osservatorio Astrofisico di Catania - Daniel Magro, University of Malta [Malta]
17:00 - 18:00	Interaction session (Breakout rooms)
18:00 - 18:10	Conference summary (Henri Mineur amphitheater / Plenary online session) - Brice Ménard (JHU)
18:10 - 19:40	Debate (Henri Mineur amphitheater / Plenary online session)
19:40 - 19:50	Concluding remarks (Henri Mineur amphitheater / Plenary online session)

Online user: 1

RSS Feed