Документ взят из кэша поисковой машины. Адрес оригинального документа : http://hea-www.harvard.edu/AstroStat/CAS2010/asiemiginowska_CAS2010aug24.pdf Дата изменения: Thu Aug 26 15:56:34 2010 Дата индексирования: Tue Oct 2 06:18:26 2012 Кодировка: Поисковые слова: кокон

Challenges and Methods for Massive Astronomical Data
mini-Workshop on Computational AstroStatistics

The work shop is sponsored by CHASC/C-BAS, NSF grants DMS 09-07185 (HU) and DMS 09-07522 (UCI), and the Chandra X-Ray Center
Au g . 2 4 -2 5 , 2 0 1 0 An e ta Si e m i g i n o wska , CfA


Small Challenges
· CHASC
Chandra AstroStatistics Collaboration

· C-BAS
California-Boston-Smithsonian AstroStatistics Collaboration

Au g . 2 4 -2 5 , 2 0 1 0

An e ta Si e m i g i n o wska , CfA


Past Challenges
1997 Meeting on "Statistical Challenges in Modern Astronomy"

Chandra*
Here are a few examples from many of workshops that have been organized by CHASC in the past: · Data Analysis Challenges in Solar & Stellar Coronal Astrophysics at AAS 199 (2002, Washington, DC) · Data Analysis Challenges in Solar and Stellar Astrophysics at AAS/SPD 2003 (2003, College Park, MD) · Current Challenges in Poisson Multi-Scale Deconvolution Methods (2003, Cambridge, MA)

* Chandra was launched on July 23, 1999

Au g . 2 4 -2 5 , 2 0 1 0

An e ta Si e m i g i n o wska , CfA


Current Challenges
· Astrostatistical efforts have focused on principled analysis of individual observations, on one or a few sources. · The new data forces us to consider combining multiple datasets and infer parameters that are common to entire populations. · Many astronomers really want to use every data point and even non-detections, but this becomes problematic for many statistical techniques.

· ....

Au g . 2 4 -2 5 , 2 0 1 0

An e ta Si e m i g i n o wska , CfA


Workshop Goals
· Explore new problems in data analysis that arise from data complexity. · Focus on problems generally considered intractable due to insufficient computational power or inefficient algorithms.
· Accounting for uncertainties in instrument calibration. · Classification, regression, and density estimations of massive data sets that may be truncated and contaminated with measurement errors and outliers. · Designing statistical emulators to efficiently approximate the output from complex astrophysical computer models and simulations, thus making statistical inference on them tractable.

· Define a path for development of new algorithms that target specific issues.
Au g . 2 4 -2 5 , 2 0 1 0 An e ta Si e m i g i n o wska , CfA


Workshop Day 1
9.30 - Noon Session 1A Moderator: Andreas Zezas (Crete) Kirk Borne (George Mason) - LSST: Informatics and Statistics Research Challenges Keith Arnaud (GSFC) - LISA: A Big Problem on a Small Data Set Brandon Kelly (SAO) - Constraining astronomical populations with truncated data sets Noon - 1:30pm : Lunch 1:30pm - 4pm : Session 1B Moderator: Paul Baines (UC Davis) Peter Freeman (CMU) - Nonlinear Data Reparametrization with Diffusion Map Joey Richards (UC Berkeley) - Real-time Classification for The Palomar Transient Factory Daryl Geller (Stony Brook) - Spherical wavelets for CMB temperature and polarization data analysis 4pm - 4:20pm : Coffee break 4:20pm - 5:30pm : Open Discussion Moderator: Vinay Kashyap (SAO)

Au g . 2 4 -2 5 , 2 0 1 0

An e ta Si e m i g i n o wska , CfA


Workshop Day 2
9:30am - Noon : Session 2A Moderator: Jeremy Drake (SAO) Paola Testa/Alisdair Davey (SAO) - Challenges in Data Distribution and Analysis with the Solar Dynamics Observatory Ashish Mahabal (CalTech) - Where statistical methods can help with Transients classification from surveys Pavlos Protopapas (SAO) : Stellar Variability Classification Using Machine Learning Noon - 1:30pm - Lunch 1:30pm - 4pm : Session 2B Moderator: Brandon Kelly (CfA) Alexander Gray (Georgia Tech) - Beyond RAM: Fast Statistical Analysis in Databases. Alex Blocker (Harvard) - Semi-parametric Robust Event Detection for Massive Time-Series Datasets Lukasz Wyrzykowski (Cambridge) - Transient classification with Gaia 4pm - 4:20pm : Coffee break 4:20pm - 5:15pm : Open Discussion Moderator: David van Dyk (UC Irvine) 5:15pm - 5:30pm : Alanna Connors (Eureka Sci) : Workshop Wrap-up

Au g . 2 4 -2 5 , 2 0 1 0

An e ta Si e m i g i n o wska , CfA


Discussion Time

Au g . 2 4 -2 5 , 2 0 1 0

An e ta Si e m i g i n o wska , CfA