The need to analyze the available large synoptic multi-band surveys drives the development of new data-analysis methods. Photometric redshift estimation is one field of application where such new methods improved the results, substantially. Up to now, the vast majority of applied redshift estimation methods have utilized photometric features. We aim to develop a method to derive probabilistic photometric redshift directly from multi-band imaging data, rendering pre-classification of objects and feature extraction obsolete. A modified version of a deep convolutional network was combined with a mixture density network. The estimates are expressed as Gaussian mixture models representing the probability density functions (PDFs) in the redshift space. In addition to the traditional scores, the continuous ranked probability score (CRPS) and the probability integral transform (PIT) were applied as performance criteria. We have adopted a feature based random forest and a plain mixture density network to compare performances on experiments with data from SDSS (DR9). We show that the proposed method is able to predict redshift PDFs independently from the type of source, for example galaxies, quasars or stars. Thereby the prediction performance is better than both presented reference methods and is comparable to results from the literature. The presented method is extremely general and allows us to solve of any kind of probabilistic regression problems based on imaging data, for example estimating metallicity or star formation rate of galaxies. This kind of methodology is tremendously important for the next generation of surveys.
Three-dimensional wide-field galaxy surveys are fundamental for cosmological studies. For higher redshifts (z > 1.0), where galaxies are too faint, quasars still trace the large-scale structure of the Universe. Since available telescope time limits spectroscopic surveys, photometric methods are efficient for estimating redshifts for many quasars. Recently, machine learning methods are increasingly successful for quasar photometric redshifts, however, they hinge on the distribution of the training set. Therefore a rigorous estimation of reliability is critical. We extracted optical and infrared…
The need for accurate photometric redshifts estimation is a topic that has fundamental importance in Astronomy, due to the necessity of efficiently obtaining redshift information without the need of spectroscopic analysis. We propose a method for determining accurate multimodal photo-z probability density functions (PDFs) using Mixture Density Networks (MDN) and Deep Convolutional Networks (DCN). A comparison with a Random Forest (RF) is performed.
We present METAPHOR (Machine-learning Estimation Tool for Accurate PHOtometric Redshifts), a method able to provide a reliable PDF for photometric galaxy redshifts estimated through empirical techniques. METAPHOR is a modular workflow, mainly based on the MLPQNA neural network as internal engine to derive photometric galaxy redshifts, but giving the possibility to easily replace MLPQNA with any other method to predict photo-z's and their PDF. We present here the results about a validation test of the workflow on the galaxies from SDSS-DR9, showing also the universality of the method by replaci…
In this work, we are trying to extent the existing photometric redshift regression models from modeling pure photometric data back to the spectra themselves. To that end, we developed a PCA that is capable of describing the input uncertainty (including missing values) in a dimensionality reduction framework. With this "spectrum generator" at hand, we are capable of treating the redshift regression problem in a fully Bayesian framework, returning a posterior distribution over the redshift. This approach allows therefore to approach the multimodal regression problem in an adequate fashion. In ad…
Photometric redshifts (photo-z) are crucial to the scientific exploitation of modern panchromatic digital surveys. In this paper we present PhotoRApToR (Photometric Research Application To Redshift): a Java/C++ based desktop application capable to solve non-linear regression and multi-variate classification problems, in particular specialized for photo-z estimation. It embeds a machine learning algorithm, namely a multilayer neural network trained by the Quasi Newton learning rule, and special tools dedicated to pre- and postprocessing data. PhotoRApToR has been successfully tested on several …
Correlating BASS DR3 catalogue with ALLWISE database, the data from optical and infrared information are obtained. The quasars from SDSS are taken as training and test samples while those from LAMOST are considered as external test sample. We propose two schemes to construct the redshift estimation models with XGBoost, CatBoost and Random forest. One scheme (namely one-step model) is to predict photometric redshifts directly based on the optimal models created by these three algorithms; the other scheme (namely two-step model) is to firstly classify the data into low- and high- redshift datase…
In the modern galaxy surveys photometric redshifts play a central role in a broad range of studies, from gravitational lensing and dark matter distribution to galaxy evolution. Using a dataset of about 25,000 galaxies from the second data release of the Kilo Degree Survey (KiDS) we obtain photometric redshifts with five different methods: (i) Random forest, (ii) Multi Layer Perceptron with Quasi Newton Algorithm, (iii) Multi Layer Perceptron with an optimization network based on the Levenberg-Marquardt learning rule, (iv) the Bayesian Photometric Redshift model (or BPZ) and (v) a classical SED…
We present a catalogue of galaxy photometric redshifts for the Sloan Digital Sky Survey (SDSS) Data Release 12. We use various supervised learning algorithms to calculate redshifts using photometric attributes on a spectroscopic training set. Two training sets are analysed in this paper. The first training set consists of 995,498 galaxies with redshifts up to $z \approx 0.8$. On the first training set, we achieve a cost function of 0.00501 and a root mean squared error value of 0.0707 using the XGBoost algorithm. We achieved an outlier rate of 2.1\% and 86.81\%, 95.83\%, 97.90\% of our data po…
We apply one of lazy learning methods named k-nearest neighbor algorithm (kNN) to estimate the photometric redshifts of quasars, based on various datasets from the Sloan Digital Sky Survey (SDSS), UKIRT Infrared Deep Sky Survey (UKIDSS) and Wide-field Infrared Survey Explorer (WISE) (the SDSS sample, the SDSS-UKIDSS sample, the SDSS-WISE sample and the SDSS-UKIDSS-WISE sample). The influence of the k value and different input patterns on the performance of kNN is discussed. kNN arrives at the best performance when k is different with a special input pattern for a special dataset. The best resu…
Based on the SDSS and SDSS-WISE quasar datasets, we put forward two schemes to estimate the photometric redshifts of quasars. Our schemes are based on the idea that the samples are firstly classified into subsamples by a classifier and then photometric redshift estimation of different subsamples is performed by a regressor. Random Forest is adopted as the core algorithm of the classifiers, while Random Forest and kNN are applied as the key algorithms of regressors. The samples are divided into two subsamples and four subsamples depending on the redshift distribution. The performance based on d…
We combine K-Nearest Neighbors (KNN) with genetic algorithm (GA) for photometric redshift estimation of quasars, short for GeneticKNN, which is a weighted KNN approach supported by GA. This approach has two improvements compared to KNN: one is the feature weighted by GA; another is that the predicted redshift is not the redshift average of K neighbors but the weighted average of median and mean of redshifts for K neighbors, i.e. $p\times z_{median} + (1-p)\times z_{mean}$. Based on the SDSS and SDSS-WISE quasar samples, we explore the performance of GeneticKNN for photometric redshift estimati…
A variety of fundamental astrophysical science topics require the determination of very accurate photometric redshifts (photo-z's). A wide plethora of methods have been developed, based either on template models fitting or on empirical explorations of the photometric parameter space. Machine learning based techniques are not explicitly dependent on the physical priors and able to produce accurate photo-z estimations within the photometric ranges derived from the spectroscopic training set. These estimates, however, are not easy to characterize in terms of a photo-z Probability Density Function…
Performance of Photometric Template Fitting for Ultra-High Redshift Galaxies
Cosmic Dawn Center (DAWN)（宇宙黎明中心）；Niels Bohr Institute, University of Copenhagen（哥本哈根大学尼尔斯·玻尔研究所）；Department of Physics and Astronomy, University of Missouri（密苏里大学物理与天文学系）；DARK Cosmology Center, University of Copenhagen（哥本哈根大学暗宇宙学中心）；California Institute of Technology（加州理工学院）
This research focuses on identifying high redshift galaxies from LABOCA(LArge APEX BOlometer CAmera) and SPIRE(The Spectral and Photometric Imaging Receiver) maps towards proto-cluster candidates initially selected from the SPT (South pole telescope) survey. Based on the Multi-Gaussian fitting algorithm, we cross-match all significant LABOCA sources at SPIRE wavelengths based on their coordinates and signal to noise ratio to derive their photometry at 250, 350, 500 and 870 $μm$. We use this information to calculate a photometric redshift for SPT sources towards cluster fields. The code was dev…
We use N-body-spectro-photometric simulations to investigate the impact of incompleteness and incorrect redshifts in spectroscopic surveys to photometric redshift training and calibration and the resulting effects on cosmological parameter estimation from weak lensing shear-shear correlations. The photometry of the simulations is modeled after the upcoming Dark Energy Survey and the spectroscopy is based on a low/intermediate resolution spectrograph with wavelength coverage of 5500Å < λ < 9500Å. The principal systematic errors that such a spectroscopic follow-up encounters are incompleteness (…
We propose a Multimodal Machine Learning method for estimating the Photometric Redshifts of quasars (PhotoRedshift-MML for short), which has long been the subject of many investigations. Our method includes two main models, i.e. the feature transformation model by multimodal representation learning, and the photometric redshift estimation model by multimodal transfer learning. The prediction accuracy of the photometric redshift was significantly improved owing to the large amount of information offered by the generated spectral features learned from photometric data via the MML. A total of 415…
In this paper we apply ideas from information theory to create a method for the design of optimal filters for photometric redshift estimation. We show the method applied to a series of simple example filters in order to motivate an intuition for how photometric redshift estimators respond to the properties of photometric passbands. We then design a realistic set of six filters covering optical wavelengths that optimize photometric redshifts for $z <= 2.3$ and $i < 25.3$. We create a simulated catalog for these optimal filters and use our filters with a photometric redshift estimation code to s…
Cosmology with Type Ia supernovae heretofore has required extensive spectroscopic follow-up to establish a redshift. Though tolerable at the present discovery rate, the next generation of ground-based all-sky survey instruments will render this approach unsustainable. Photometry-based redshift determination is a viable alternative, but introduces non-negligible errors that ultimately degrade the ability to discriminate between competing cosmologies. We present a strictly template-based photometric redshift estimator and compute redshift reconstruction errors in the presence of photometry and s…
We present a Bayesian hierarchical framework to analyze photometric galaxy survey data with stellar population synthesis (SPS) models. Our method couples robust modeling of spectral energy distributions with a population model and a noise model to characterize the statistical properties of the galaxy populations and real observations, respectively. By self-consistently inferring all model parameters, from high-level hyper-parameters to SPS parameters of individual galaxies, one can separate sources of bias and uncertainty in the data.We demonstrate the strengths and flexibility of this approac…
We present catalogs of calibrated photometry and spectroscopic redshifts in the Extended Groth Strip, intended for studies of photometric redshifts (photo-z's). The data includes ugriz photometry from CFHTLS and Y-band photometry from the Subaru Suprime camera, as well as spectroscopic redshifts from the DEEP2, DEEP3 and 3D-HST surveys. These catalogs incorporate corrections to produce effectively matched-aperture photometry across all bands, based upon object size information available in the catalog and Moffat profile point spread function fits. We test this catalog with a simple machine lea…
Machine learning (ML) is a standard approach for estimating the redshifts of galaxies when only photometric information is available. ML photo-z solutions have traditionally ignored the morphological information available in galaxy images or partly included it in the form of hand-crafted features, with mixed results. We train a morphology-aware photometric redshift machine using modern deep learning tools. It uses a custom architecture that jointly trains on galaxy fluxes, colors and images. Galaxy-integrated quantities are fed to a Multi-Layer Perceptron (MLP) branch while images are fed to a…
In this study, we propose a three-stage training approach of neural networks for both photometric redshift estimation of galaxies and detection of out-of-distribution (OOD) objects. Our approach comprises supervised and unsupervised learning, which enables using unlabeled (UL) data for OOD detection in training the networks. Employing the UL data, which is the dataset most similar to the real-world data, ensures a reliable usage of the trained model in practice. We quantitatively assess the model performance of photometric redshift estimation and OOD detection using in-distribution (ID) galaxi…
We present results of using individual galaxies' redshift probability information derived from a photometric redshift (photo-z) algorithm, SPIDERz, to identify potential catastrophic outliers in photometric redshift determinations. By using two test data sets comprised of COSMOS multi-band photometry spanning a wide redshift range (0<z<4) matched with reliable spectroscopic or other redshift determinations we explore the efficacy of a novel method to flag potential catastrophic outliers in an analysis which relies on accurate photometric redshifts. SPIDERz is a custom support vector machine cl…
Machine learning techniques, specifically the k-nearest neighbour algorithm applied to optical band colours, have had some success in predicting photometric redshifts of quasi-stellar objects (QSOs): Although the mean of differences between the spectroscopic and photometric redshifts is close to zero, the distribution of these differences remains wide and distinctly non-Gaussian. As per our previous empirical estimate of photometric redshifts, we find that the predictions can be significantly improved by adding colours from other wavebands, namely the near-infrared and ultraviolet. Self-testin…
Accurate photometric redshifts for large samples of galaxies are among the main products of modern multiband digital surveys. Over the last decade, the Sloan Digital Sky Survey (SDSS) has become a sort of benchmark against which to test the various methods. We present an application of a new method to the estimation of photometric redshifts for the galaxies in the SDSS Data Release 9 (SDSS-DR9). Photometric redshifts for more than 143 million galaxies were produced and made available at the URL: http://dame.dsf.unina.it/catalog/DR9PHOTOZ/. The MLPQNA (Multi Layer Perceptron with Quasi Newton A…
We propose a new method to estimate the photometric redshift of galaxies by using the full galaxy image in each measured band. This method draws from the latest techniques and advances in machine learning, in particular Deep Neural Networks. We pass the entire multi-band galaxy image into the machine learning architecture to obtain a redshift estimate that is competitive with the best existing standard machine learning techniques. The standard techniques estimate redshifts using post-processed features, such as magnitudes and colours, which are extracted from the galaxy images and are deemed t…
University of Pittsburgh（匹兹堡大学）；Federal University of São Carlos（圣卡洛斯联邦大学）；Carnegie Mellon University（卡内基梅隆大学）；Argonne National Laboratory（阿贡国家实验室）；Ruhr-University Bochum（波鸿鲁尔大学）；German Centre for Cosmological Lensing（德国宇宙透镜中心）
Many astrophysical analyses depend on estimates of redshifts (a proxy for distance) determined from photometric (i.e., imaging) data alone. Inaccurate estimates of photometric redshift uncertainties can result in large systematic errors. However, probability distribution outputs from many photometric redshift methods do not follow the frequentist definition of a Probability Density Function (PDF) for redshift -- i.e., the fraction of times the true redshift falls between two limits $z_{1}$ and $z_{2}$ should be equal to the integral of the PDF between these limits. Previous works have used the…
Measuring distances of cosmological sources such as galaxies, stars and quasars plays an increasingly critical role in modern cosmology. Obtaining the optical spectrum and consequently calculating the redshift as a distance indicator could instantly classify these objects. As long as spectroscopic observations are not available for many galaxies and the process of measuring the redshift is time-consuming and infeasible for large samples, machine learning (ML) approaches could be applied to determine the redshifts of galaxies from different features including their photometric colors. In this p…
In this paper we explore the applicability of the unsupervised machine learning technique of Self Organizing Maps (SOM) to estimate galaxy photometric redshift probability density functions (PDFs). This technique takes a spectroscopic training set, and maps the photometric attributes, but not the redshifts, to a two dimensional surface by using a process of competitive learning where neurons compete to more closely resemble the training data multidimensional space. The key feature of a SOM is that it retains the topology of the input set, revealing correlations between the attributes that are …
