Sacrificing information for the greater good: how to select photometric bands for optimal accuracy
arXiv:1511.05424 · doi:10.1093/mnras/stw2476
Abstract
Large-scale surveys make huge amounts of photometric data available. Because of the sheer amount of objects, spectral data cannot be obtained for all of them. Therefore it is important to devise techniques for reliably estimating physical properties of objects from photometric information alone. These estimates are needed to automatically identify interesting objects worth a follow-up investigation as well as to produce the required data for a statistical analysis of the space covered by a survey. We argue that machine learning techniques are suitable to compute these estimates accurately and efficiently. This study promotes a feature selection algorithm, which selects the most informative magnitudes and colours for a given task of estimating physical quantities from photometric data alone. Using k nearest neighbours regression, a well-known non-parametric machine learning method, we show that using the found features significantly increases the accuracy of the estimations compared to using standard features and standard methods. We illustrate the usefulness of the approach by estimating specific star formation rates (sSFRs) and redshifts (photo-z's) using only the broad-band photometry from the Sloan Digital Sky Survey (SDSS). For estimating sSFRs, we demonstrate that our method produces better estimates than traditional spectral energy distribution (SED) fitting. For estimating photo-z's, we show that our method produces more accurate photo-z's than the method employed by SDSS. The study highlights the general importance of performing proper model selection to improve the results of machine learning systems and how feature selection can provide insights into the predictive relevance of particular input features.
arXiv comment20 pages, 13 figures, 3 tables. v2: Significant changes to match revised version, including new photo-z estimation experiment
References in corpus (61)
- Scikit-learn: Machine Learning in Python
- The NumPy array: a structure for efficient numerical computation
- The Sloan Digital Sky Survey: Technical Summary
- Star Formation in Galaxies Along the Hubble Sequence
- The Seventh Data Release of the Sloan Digital Sky Survey
- The physical properties of star forming galaxies in the low redshift universe
- Star Formation in the Milky Way and Nearby Galaxies
- The Galaxy Evolution Explorer: A Space Ultraviolet Survey Mission
- Stellar Masses and Star Formation Histories for 10^5 Galaxies from the Sloan Digital Sky Survey
- EAZY: A Fast, Public Photometric Redshift Code
- Accurate photometric redshifts for the CFHT Legacy Survey calibrated using the VIMOS VLT Deep Survey
- UV Star Formation Rates in the Local Universe
- A simple model to interpret the ultraviolet, optical and infrared emission from galaxies
- The ages and metallicities of galaxies in the local universe
- Bayesian photometric redshift estimation
- Analysis of galaxy SEDs from far-UV to far-IR with CIGALE: Studying a SINGS test sample
- Modeling the Panchromatic Spectral Energy Distributions of Galaxies
- Detection of quiescent galaxies in a bicolor sequence from z=0-2
- Galaxy evolution in groups and clusters: satellite star formation histories and quenching timescales in a hierarchical Universe
- Galaxy evolution in groups and clusters: star formation rates, red sequence fractions, and the persistent bimodality
- ANNz: estimating photometric redshifts using artificial neural networks
- On Star Formation Rates and Star Formation Histories of Galaxies out to z ~ 3
- A Critical Assessment of Photometric Redshift Methods: A CANDELS Investigation
- The Clustering of Luminous Red Galaxies in the Sloan Digital Sky Survey Imaging Data
- Fitting the integrated Spectral Energy Distributions of Galaxies
- Star formation rates and masses of z ~ 2 galaxies from multicolour photometry
- TPZ : Photometric redshift PDFs and ancillary information by using prediction trees and random forests
- PHAT: PHoto-z Accuracy Testing
- A CANDELS - 3D-HST Synergy: Resolved Star Formation Patterns at 0.7 < z < 1.5
- Encoding the infrared excess (IRX) in the NUVrK color diagram for star-forming galaxies
- New Constraints on the Star Formation Histories and Dust Attenuation of Galaxies in the Local Universe from GALEX
- Photometric redshift analysis in the Dark Energy Survey Science Verification data
- A census of metals and baryons in stars in the local Universe
- Clustering of Sloan Digital Sky Survey III Photometric Luminous Galaxies: The Measurement, Systematics and Cosmological Implications
- GAMA/H-ATLAS: A meta-analysis of SFR indicators - comprehensive measures of the SFR-M* relation and Cosmic Star Formation History at z < 0.4
- Taking Care of Business in a Flash: Constraining the Timescale for Low-Mass Satellite Quenching with ELVIS
- Rapid Environmental Quenching of Satellite Dwarf Galaxies in the Local Group
- Evidence for a change in the dominant satellite galaxy quenching mechanism at z=1
- The surprising inefficiency of dwarf satellite quenching
- Spectral Templates from Multicolor Redshift Surveys
- Creating Spectral Templates from Multicolor Redshift Surveys
- A Comparison of Six Photometric Redshift Methods Applied to 1.5 Million Luminous Red Galaxies
- Star Formation, Metallicity and Dust Properties Derived from the SAPM Galaxy Survey Spectra
- On the importance of using appropriate spectral models to derive physical properties of galaxies at 0.7<z<2.8
- Unsupervised self-organised mapping: a versatile empirical tool for object selection, classification and redshift estimation in large surveys
- A Sparse Gaussian Process Framework for Photometric Redshift Estimation
- Feature importance for machine learning redshifts applied to SDSS galaxies
- Clustering of Photometric Luminous Red Galaxies II: Cosmological Implications from the Baryon Acoustic Scale
- Constraints on the star formation histories of galaxies from z~1 to z~0
- An analysis of feature relevance in the classification of astronomical transients with machine learning methods
- Deriving star formation histories from photometry using energy balance spectral energy distribution modelling
- Accelerating Nearest Neighbor Search on Manycore Systems
- Automated classification of Hipparcos unsolved variables
- Machine-assisted discovery of relationships in astronomy
- Implementation of a Parallel Tree Method on a GPU
- A First Look at creating mock catalogs with machine learning techniques
- Accurate parameter estimation for star formation history in galaxies using SDSS spectra
- Finding New High-Redshift Quasars by Asking the Neighbours
- Towards a library of synthetic galaxy spectra and preliminary results of classification and parametrization of unresolved galaxies for Gaia
- An Active Instance-based Machine Learning method for Stellar Population Studies
- Estimating Spectroscopic Redshifts by Using k Nearest Neighbors Regression I. Description of Method and Analysis
Cited by in corpus (13)
- Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
- Big Universe, Big Data: Machine Learning and Image Analysis for Astronomy
- On the realistic validation of photometric redshifts, or why Teddy will never be Happy
- A machine learning approach to galaxy properties: joint redshift-stellar mass probability distributions with Random Forest
- Horizon-AGN virtual observatory -- 2: Template-free estimates of galaxy properties from colours
- Star Formation Rates for photometric samples of galaxies using machine learning methods
- Predicting star formation properties of galaxies using deep learning
- {\sc mirkwood:} Fast and Accurate SED Modeling Using Machine Learning
- Photometric redshifts with machine learning, lights and shadows on a complex data science use case
- Comparison of Observed Galaxy Properties with Semianalytic Model Predictions using Machine Learning
- Foreword to the Focus Issue on Machine Learning in Astronomy and Astrophysics
- Euclid preparation. LXVIII. Extracting physical parameters from galaxies with machine learning
- A Value-added Physical Properties Catalog for Low-redshift Galaxies from DESI Legacy Imaging Surveys DR10