Results from the Supernova Photometric Classification Challenge

Abstract

We have publicly released a blinded mix of simulated SNe, with types (Ia, Ib, Ic, II) selected in proportion to their expected rate. The simulation is realized in the griz filters of the Dark Energy Survey (DES) with realistic observing conditions (sky noise, point spread function and atmospheric transparency) based on years of recorded conditions at the DES site. Simulations of non-Ia type SNe are based on spectroscopically confirmed light curves that include unpublished non-Ia samples donated from the Carnegie Supernova Project (CSP), the Supernova Legacy Survey (SNLS), and the Sloan Digital Sky Survey-II (SDSS-II). We challenge scientists to run their classification algorithms and report a type for each SN. A spectroscopically confirmed subset is provided for training. The goals of this challenge are to (1) learn the relative strengths and weaknesses of the different classification algorithms, (2) use the results to improve classification algorithms, and ( 3) understand what spectroscopically confirmed sub-sets are needed to properly train these algorithms. The challenge is available at www.hep.anl.gov/SNchallenge, and the due date for classifications is May 1, 2010.

Observational probes of cosmic acceleration

“…One can also apply data quality and other selection cuts before the spectroscopic observations to reduce the total number of spectra required, though one must be careful not to let biases creep in at this stage. With good photometric monitoring and with subsequent spectroscopic redshifts of apparent hosts, Kessler et al (2010) find that they can identify Type Ia SNe with 70% to 90% confidence from the LCS and color alone, and Bernstein et al (2012) forecast Type Ia purity as high as 98% for DES-like photometric observations. A moderate amount of real-time supernova spectroscopy may then suffice to assess efficiency and biases.…”

The discovery and classification of 16 supernovae at high redshifts in ELAIS-S1

“…Kessler et al (2010a) try to remedy this by offering a test sample of SN light curves and a framework in which different typing codes can be compared. Currently, our code is written specifically to deal with the data from SVISS, trying the code out on the test sample is thus outside the scope of this paper.…”

“…Currently, our code is written specifically to deal with the data from SVISS, trying the code out on the test sample is thus outside the scope of this paper. Nevertheless, we can compare the typing quality estimated from our Monte-Carlosimulated light curves with the typing qualities reported in Kessler et al (2010a). The Ia efficiency, which in our paper is equal to (N app (T N) − N TN→CC )/N tot (T N), and Ia pureness, (N app (T N) − N TN→CC )/(N app (T N) − N TN→CC + N CC→TN ) in our notation, are two of the concepts used to describe the quality in Kessler et al (2010a).…”

“…Nevertheless, we can compare the typing quality estimated from our Monte-Carlosimulated light curves with the typing qualities reported in Kessler et al (2010a). The Ia efficiency, which in our paper is equal to (N app (T N) − N TN→CC )/N tot (T N), and Ia pureness, (N app (T N) − N TN→CC )/(N app (T N) − N TN→CC + N CC→TN ) in our notation, are two of the concepts used to describe the quality in Kessler et al (2010a). Our overall efficiency and pureness values for the thermonuclear SNe, 0.97 and 0.75, respectively, are comparable to the results obtained by the other typing codes, but with the caveat that we did not use the same test sample of supernovae as these authors.…”

“…With this method the templates are assigned an uncertainty that enters into the likelihood calculations, which improves the classification quality of SNe with non-standard light curves. Recently, Kessler et al (2010a) presented the results of a supernova classification challenge, where a common sample of supernova light curves was typed by different codes. Their results indicate that Bayesian typing codes are competitive and give reliable type determinations.…”