Applications of Gaussian Processes to Supernova Data

Details Applications of Gaussian Processes to Supernova Data Applications of Gaussian Processes to Supernova Data

Jan 2012

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2012aas...21924222t&link_type=abstract

American Astronomical Society, AAS Meeting #219, #242.22

Statistics

Computation

Scientific paper

We demonstrate the use of Gaussian processes in problems relevant to Type Ia supernova cosmology experiments and the analysis of supernovae in general. Gaussian processes are a powerful statistical approach that generalizes the concept of probability distributions over random variables to functions. Nonlinear regression, smoothing, and machine classification problems are target applications of Gaussian processes. Areas where Gaussian processes may be an interesting solution in Type Ia supernova cosmology are: principled construction of spectroscopic surface templates, robust extraction of spectral feature measurements, and light curve fitting/modeling. We describe our high-performance computer framework that scales to data sets of interest to current and near-term cosmology experiments, describe computational challenges in the implementation (and their resolution), and show example results using data from the Nearby Supernova Factory and simulations from the Dark Energy Survey.

Affiliated with

Also associated with

No associations

Rating

Applications of Gaussian Processes to Supernova Data does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Applications of Gaussian Processes to Supernova Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Applications of Gaussian Processes to Supernova Data will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1578568