Astro@Stats 2017: Sino-Italian Workshop on Astrostatistics

Europe/Rome
SC140 (Department of Statistical Sciences)

SC140

Department of Statistical Sciences

Via C. Battisti, 241 35121 Padova, Italy
Alessandra R Brazzale (University of Padova), Denis Bastieri (University of Padova and INFN Padova), Giorgio Picci (University of Padova)
Description

Modern astronomy and cosmology heavily rely on statistical methods. At the same time, the recent explosion of available data, which for instance results from current and future astronomical sky surveys, provides unique challenges, and opportunities, to statistics. Several statistical challenges appearing in astrophysical research also arise in other fields of physics, such as particle physics.

The aim of this workshop is to foster the interchange of ideas among astronomers, physicists and statisticians working at, and with, the University of Padova. The event has three sessions dedicated to academic talks by senior researchers, a junior speed session and a poster session open to a limited number of participants, who will be notified after the abstract review has been completed.

Registration to the workshop is open, though because of the limited places available it will be limited to 80 participants. A registration fee of 25 Euro is required. Registration fee waivers for students are available and will be distributed at the discretion of the scientific committee.

For further information send an email to astro2017@stat.unipd.it

You can also download the poster of the meeting.



Co-organized with With the endorsement of
Dept. of Physics and Astronomy University of Padova
Confucius Institute Padova Guangzhou University

 

Endorsing institutions      



Supported by the Confucius Institute at the University of Padova
    • 08:30 09:00
      Registration 30m SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy
    • 09:00 09:30
      Welcome SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy
      • 09:00
        Welcome by Vice Rector for International Relations 10m
        Prof. Alessandro Paccagnella, Vice Rector for International Relations
        Speaker: Alessandro Paccagnella (University of Padova)
      • 09:10
        Welcome by Chair of the Department of Statistical Sciences 10m
        Prof. Tommaso di Fonzo, Chair of the Department of Statistical Sciences
        Speaker: Tommaso Di Fonzo (University of Padova)
      • 09:20
        Welcome by Chair of the Department of Physics and Astronomy 10m
        Prof. Francesca Soramel, Chair of the Department of Physics and Astronomy
        Speaker: Francesca Soramel (University of Padova)
    • 09:30 11:00
      J.C.F. Gauss: Astrophysics meets Statistics SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy

      Astrophysics meets Statistics

      Convener: Alessandra R. Brazzale (University of Padova)
      • 09:30
        Investigating the Cosmic Web with Topological Data Analysis 30m
        Data exhibiting complicated spatial structures are common in many areas of science (e.g. cosmology, biology), but can be difficult to analyze. Persistent homology is a popular approach within the area of Topological Data Analysis (TDA) that offers a new way to represent, visualize, and interpret complex data by extracting topological features, which can be used to infer properties of the underlying structures. In particular, TDA may be useful for analyzing the large-scale structure (LSS) of the Universe, which is an intricate and spatially complex web of matter. In order to understand the physics of the Universe, theoretical and computational cosmologists develop large-scale simulations that allow for visualizing and analyzing the LSS under varying physical assumptions. Each point in the 3D data set represents a galaxy or a cluster of galaxies, and topological summaries ("persistent diagrams") can be obtained summarizing the different ordered holes in the data (e.g. connected components, loops, voids). The topological summaries are interesting and informative descriptors of the Universe on their own, but hypothesis tests using the topological summaries would provide a way to make more rigorous comparisons of LSS under different theoretical models. For example, the received cosmological model has cold dark matter (CDM); however, while the case is strong for CDM, there are some observational inconsistencies with this theory. Another possibility is warm dark matter (WDM). It is of interest to see if a CDM Universe and WDM Universe produce LSS that is topologically distinct. We present several possible test statistics for two-sample hypothesis tests using the topological summaries, carry out a simulation study to investigate the suitableness of the proposed test statistics using simulated data from a variation of the Voronoi foam model, and finally we apply the proposed inference framework to WDM vs. CDM cosmological simulation data.
        Speaker: Jessi Cisewski-Kehe (Yale University)
      • 10:00
        Quasars: the Beacons of the Universe and the Challenge for Machine Learning 30m
        Speaker: Maria Süveges (Max Planck Institute for Astronomy Heidelberg)
      • 10:30
        Photometric Classification of Supernova with a Biased Training Set 30m
        The expansion of the Universe can be studied via a Bayesian model that relates the difference between the apparent and intrinsic brightnesses of objects to their distance which in turn depends on parameters that describe this expansion. While apparent brightness can be readily measured, intrinsic brightness can only be obtained for certain objects. Type Ia Supernova (SNIa) occur when material accreting onto a white dwarf triggers a powerful supernova explosion. Because this occurs only in a particular physical scenario, we can estimate the intrinsic brightness of SNIa. To take advantage of this, however, SNIa must be precisely classified using (low resolution) photometric data and a biased training set. We use Gaussian Processes to account for irregular observation times and diffusion maps to identify features for a random forest classifier. To account for bias in the overall training set, we use propensity scores to form homogeneous groups where the training subsets are more representative. Finally we enrich the training sets by probabilistically generating synthetic data. In this way we are able to identify SNIa nearly as well as we would with an unbiased training set.
        Speaker: David Van Dyk (Imperial College London)
    • 11:00 11:30
      Coffee Break 30m "B. Colombo" Library

      "B. Colombo" Library

      Department of Statistical Sciences

      Tea and coffee break

    • 11:30 12:30
      G. Galilei: Physics meets Statistics SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy

      Physics meets Statistics

      Convener: Denis Bastieri (PD)
      • 11:30
        New Statistical Analyses for the Neutrino Mass Hierarchy Determination 30m
        We preliminary focus on some aspects of the statistical analyses in particle physics. Then, two new analyses on the neutrino mass hierachy (MA) will be presented. The first focuses on the determination of MA with neutrino from accelerator beams, while the second one with antineutrino from reactor plants.
        Speaker: Luca Stanco (INFN Padova)
      • 12:00
        Nonparametric Semi-Supervised Classification with Application to Signal Detection in High Energy Physics 30m
        Since the early Sixties, the Standard Model has represented the state of the art in High Energy Physics. It describes how the fundamental particles interact with each others and with the forces between them, giving rise to the matter in the universe. Despite its empirical confirmations, there are indications that the Standard Model does itself not complete our understanding of the universe. Model independent search aims to explain the shortcomings of this theory by empirically looking for any possible signal which behaves as a deviation from the background process, representing, in turn, the known physics. From a statistical perspective, this problem can be in principle formulated within an unsupervised framework of clustering. However, while the the signal, if present, is unknown, the background process is always present and well-known, so that a virtually infinite sample of data can be simulated from the latter process with Montecarlo techinques. Hence, available data have two different sources: an unlabelled sample which might include observations from both the processes, and an additional labelled, sample from the background only. A semisupervised approach can be particularly suitable in this context, for discriminating the two class labels; semisupervised classification techniques lie between unsupervised and supervised ones, sharing some characteristics of both the approaches. In this work we propose a procedure where additional information, available on the background, is integrated within a nonparametric clustering framework to detect deviations from known physics. Also, we propose a variable selection procedure that allows to work on a reduced subspace. The effectiveness of the whole methodology is shown via its application on a set of data related to a simulated experiment of a proton-proton collision.
        Speaker: Alessandro Casa (University of Padova)
    • 12:30 14:30
      Lunch 2h
    • 14:30 16:00
      孔子: Confucius - China meets Italy SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy

      Confucius: China meets Italy

      Convener: Chongqi Zhang (Guangzhou University)
      • 14:30
        TBA 30m
        Speaker: Junhui Fan (Guangzhou University Center for Astrophysics)
      • 15:00
        New Statistical Analyses for the Neutrino Mass Hierarchy Determination 30m
        We preliminary focus on some aspects of the statistical analyses in particle physics. Then, two new analyses on the neutrino mass hierachy (MA) will be presented. The first focuses on the determination of MA with neutrino from accelerator beams, while the second one with antineutrino from reactor plants.
        Speaker: Emilio Ciuffoli (Chinese Academy of Sciences)
      • 15:30
        State-Space Modeling and Identification of Large-Dimensional Multivariate Time-Series 30m
        I shall first discuss traditional ARMA models and the difficulties one has to face for the estimation of such models when dealing with high-dimensional multivariate time series. I shall then illustrate State Space and Dynamic Factor Analysis models which have been recently introduced in the literature. Estimation techniques which are reliable and numerically robust exist which have been used successfully in many applications.
        Speaker: Giorgio Picci (University of Padova)
    • 16:00 16:30
      Coffee Break 30m "B. Colombo" Library

      "B. Colombo" Library

      Department of Statistical Sciences

    • 16:30 17:00
      A. Dent: Junior Speed Session SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy

      Junior speed session

      Convener: Lorenzo Maragoni
      • 16:30
        Covariance-Consistent Vector ARMA Modeling 5m
        Speaker: Zhu Bin (University of Padova)
      • 16:35
        Spatial and Time Clustering of the High-Energy Photons Collected by the Fermi LAT 5m
        Speaker: Denise Costantin (Guangzhou University)
      • 16:40
        Finite Dirichlet Mixture Modeling for Classification and Detection of New Classes of Variable Stars 5m
        Speaker: Prince John (University of Padova)
      • 16:45
        Model Independent Searches for New Physics via Parametric Anomaly Detection 5m
        Speaker: Grzegorz M Kotkowski (University of Padova)
      • 16:50
        Line-Profile Variations in Radial-Velocity Measurements Using a Skew Normal Distribution 5m
        Speaker: Umberto Simola (University of Padova & Yale University)
      • 16:55
        Searching for Gamma Ray Sources in the Extra-Galactic Space: A Statistical Analysis of the Fermi LAT Data 5m
        Speaker: Andrea Sottosanti (University of Padova)
    • 17:00 17:30
      G. Galilei SC140

      SC140

      Department of Statistical Sciences

      Via C. Battisti, 241 35121 Padova, Italy

      Physics meets Statistics

      • 17:00
        Extraordinary Claims: the 0.000029% Solution 30m
        Extraordinary claims require extraordinary evidence. The p < 0.000029% criterion used in HEP and astro-HEP does not appear adequate to cover all experimental situations. The seminar will start with a short history of anomalies found in HEP data and their resolution, and then focus on the statistical problem of defining a proper discovery level for new phenomena and on the non-trivial issues it entails. The seminar will be in part based on this recent article: [Extraordinary claims: the 0.000029% solution](http://www.epj-conferences.org/articles/epjconf/pdf/2015/14/epjconf_icnfp2014_02003.pdf)
        Speaker: Tommaso Dorigo (University of Padova &amp; INFN Padova)
    • 17:30 18:30
      Poster session: and closing reception "B. Colombo" Library

      "B. Colombo" Library

      Department of Statistical Sciences

      and closing Reception