Rapid, machine-learned resource allocation: Application to high-redshift gamma-ray burst follow-up

A. N. Morgan, James Long, Joseph W. Richards, Tamara Broderick, Nathaniel Butler, Joshua S. Bloom

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

As the number of observed gamma-ray bursts (GRBs) continues to grow, follow-up resources need to be used more efficiently in order to maximize science output from limited telescope time. As such, it is becoming increasingly important to rapidly identify bursts of interest as soon as possible after the event, before the afterglows fade beyond detectability. Studying the most distant (highest redshift) events, for instance, remains a primary goal for many in the field. Here, we present our Random Forest Automated Triage Estimator for GRB redshifts (RATE GRB-z ) for rapid identification of high-redshift candidates using early-time metrics from the three telescopes onboard Swift. While the basic RATE methodology is generalizable to a number of resource allocation problems, here we demonstrate its utility for telescope-constrained follow-up efforts with the primary goal to identify and study high-z GRBs. For each new GRB, RATE GRB-z provides a recommendation - based on the available telescope time - of whether the event warrants additional follow-up resources. We train RATE GRB-z using a set consisting of 135 Swift bursts with known redshifts, only 18 of which are z > 4. Cross-validated performance metrics on these training data suggest that 56% of high-z bursts can be captured from following up the top 20% of the ranked candidates, and 84% of high-z bursts are identified after following up the top 40% of candidates. We further use the method to rank 200 + Swift bursts with unknown redshifts according to their likelihood of being high-z.

Original languageEnglish (US)
Article number170
JournalAstrophysical Journal
Volume746
Issue number2
DOIs
StatePublished - Feb 20 2012

Fingerprint

resource allocation
gamma ray bursts
bursts
telescopes
resource
train
resources
methodology
afterglows
recommendations
estimators
education
output

Keywords

  • gamma-ray burst: general
  • methods: data analysis
  • methods: statistical

ASJC Scopus subject areas

  • Space and Planetary Science
  • Astronomy and Astrophysics

Cite this

Rapid, machine-learned resource allocation : Application to high-redshift gamma-ray burst follow-up. / Morgan, A. N.; Long, James; Richards, Joseph W.; Broderick, Tamara; Butler, Nathaniel; Bloom, Joshua S.

In: Astrophysical Journal, Vol. 746, No. 2, 170, 20.02.2012.

Research output: Contribution to journalArticle

Morgan, A. N. ; Long, James ; Richards, Joseph W. ; Broderick, Tamara ; Butler, Nathaniel ; Bloom, Joshua S. / Rapid, machine-learned resource allocation : Application to high-redshift gamma-ray burst follow-up. In: Astrophysical Journal. 2012 ; Vol. 746, No. 2.
@article{2be555212175487f80935059e2a90c4a,
title = "Rapid, machine-learned resource allocation: Application to high-redshift gamma-ray burst follow-up",
abstract = "As the number of observed gamma-ray bursts (GRBs) continues to grow, follow-up resources need to be used more efficiently in order to maximize science output from limited telescope time. As such, it is becoming increasingly important to rapidly identify bursts of interest as soon as possible after the event, before the afterglows fade beyond detectability. Studying the most distant (highest redshift) events, for instance, remains a primary goal for many in the field. Here, we present our Random Forest Automated Triage Estimator for GRB redshifts (RATE GRB-z ) for rapid identification of high-redshift candidates using early-time metrics from the three telescopes onboard Swift. While the basic RATE methodology is generalizable to a number of resource allocation problems, here we demonstrate its utility for telescope-constrained follow-up efforts with the primary goal to identify and study high-z GRBs. For each new GRB, RATE GRB-z provides a recommendation - based on the available telescope time - of whether the event warrants additional follow-up resources. We train RATE GRB-z using a set consisting of 135 Swift bursts with known redshifts, only 18 of which are z > 4. Cross-validated performance metrics on these training data suggest that 56{\%} of high-z bursts can be captured from following up the top 20{\%} of the ranked candidates, and 84{\%} of high-z bursts are identified after following up the top 40{\%} of candidates. We further use the method to rank 200 + Swift bursts with unknown redshifts according to their likelihood of being high-z.",
keywords = "gamma-ray burst: general, methods: data analysis, methods: statistical",
author = "Morgan, {A. N.} and James Long and Richards, {Joseph W.} and Tamara Broderick and Nathaniel Butler and Bloom, {Joshua S.}",
year = "2012",
month = "2",
day = "20",
doi = "10.1088/0004-637X/746/2/170",
language = "English (US)",
volume = "746",
journal = "Astrophysical Journal",
issn = "0004-637X",
publisher = "IOP Publishing Ltd.",
number = "2",

}

TY - JOUR

T1 - Rapid, machine-learned resource allocation

T2 - Application to high-redshift gamma-ray burst follow-up

AU - Morgan, A. N.

AU - Long, James

AU - Richards, Joseph W.

AU - Broderick, Tamara

AU - Butler, Nathaniel

AU - Bloom, Joshua S.

PY - 2012/2/20

Y1 - 2012/2/20

N2 - As the number of observed gamma-ray bursts (GRBs) continues to grow, follow-up resources need to be used more efficiently in order to maximize science output from limited telescope time. As such, it is becoming increasingly important to rapidly identify bursts of interest as soon as possible after the event, before the afterglows fade beyond detectability. Studying the most distant (highest redshift) events, for instance, remains a primary goal for many in the field. Here, we present our Random Forest Automated Triage Estimator for GRB redshifts (RATE GRB-z ) for rapid identification of high-redshift candidates using early-time metrics from the three telescopes onboard Swift. While the basic RATE methodology is generalizable to a number of resource allocation problems, here we demonstrate its utility for telescope-constrained follow-up efforts with the primary goal to identify and study high-z GRBs. For each new GRB, RATE GRB-z provides a recommendation - based on the available telescope time - of whether the event warrants additional follow-up resources. We train RATE GRB-z using a set consisting of 135 Swift bursts with known redshifts, only 18 of which are z > 4. Cross-validated performance metrics on these training data suggest that 56% of high-z bursts can be captured from following up the top 20% of the ranked candidates, and 84% of high-z bursts are identified after following up the top 40% of candidates. We further use the method to rank 200 + Swift bursts with unknown redshifts according to their likelihood of being high-z.

AB - As the number of observed gamma-ray bursts (GRBs) continues to grow, follow-up resources need to be used more efficiently in order to maximize science output from limited telescope time. As such, it is becoming increasingly important to rapidly identify bursts of interest as soon as possible after the event, before the afterglows fade beyond detectability. Studying the most distant (highest redshift) events, for instance, remains a primary goal for many in the field. Here, we present our Random Forest Automated Triage Estimator for GRB redshifts (RATE GRB-z ) for rapid identification of high-redshift candidates using early-time metrics from the three telescopes onboard Swift. While the basic RATE methodology is generalizable to a number of resource allocation problems, here we demonstrate its utility for telescope-constrained follow-up efforts with the primary goal to identify and study high-z GRBs. For each new GRB, RATE GRB-z provides a recommendation - based on the available telescope time - of whether the event warrants additional follow-up resources. We train RATE GRB-z using a set consisting of 135 Swift bursts with known redshifts, only 18 of which are z > 4. Cross-validated performance metrics on these training data suggest that 56% of high-z bursts can be captured from following up the top 20% of the ranked candidates, and 84% of high-z bursts are identified after following up the top 40% of candidates. We further use the method to rank 200 + Swift bursts with unknown redshifts according to their likelihood of being high-z.

KW - gamma-ray burst: general

KW - methods: data analysis

KW - methods: statistical

UR - http://www.scopus.com/inward/record.url?scp=84856843981&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84856843981&partnerID=8YFLogxK

U2 - 10.1088/0004-637X/746/2/170

DO - 10.1088/0004-637X/746/2/170

M3 - Article

AN - SCOPUS:84856843981

VL - 746

JO - Astrophysical Journal

JF - Astrophysical Journal

SN - 0004-637X

IS - 2

M1 - 170

ER -