Estimating Photometric Redshifts Using Support Vector Machines | Zendy

Yogesh Wadadekar | Zendy

AI Assistant Blog Pricing

Home ZAIA Blog

Open Access

Estimating Photometric Redshifts Using Support Vector Machines

Author(s) -

Yogesh Wadadekar

Publication year - 2005

Publication title -

publications of the astronomical society of the pacific

Language(s) - English

Resource type - Journals

SCImago Journal Rank - 2.294

H-Index - 172

eISSN - 1538-3873

pISSN - 0004-6280

DOI - 10.1086/427710

Subject(s) - photometric redshift , redshift , galaxy , support vector machine , surface brightness , sky , computer science , brightness , artificial neural network , spectral energy distribution , astrophysics , kernel (algebra) , pattern recognition (psychology) , artificial intelligence , physics , astronomy , mathematics , combinatorics

We present a new approach to obtaining photometric redshifts using a kernellearning technique called Support Vector Machines (SVMs). Unlike traditionalspectral energy distribution fitting, this technique requires a large andrepresentative training set. When one is available, however, it is likely toproduce results that are comparable to the best obtained using template fittingand artificial neural networks. Additional photometric parameters such asmorphology, size and surface brightness can be easily incorporated. Thetechnique is demonstrated using samples of galaxies from the Sloan Digital SkySurvey Data Release 2 and the hybrid galaxy formation code GalICS. The RMSerror in redshift estimation is $<0.03$ for both samples. The strengths andlimitations of the technique are assessed.Comment: 10 pages, 3 figures, to appear in the PASP, minor typos fixed to make consistent with published versio

The content you want is available to Zendy users.

Already have an account? Click here to sign in.

Having issues? You can contact us here

Accelerating Research