Abstract:
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 to be salient by the user. This new method removes the user from the photometric redshift estimation pipeline. However we do note that Deep Neural Networks require many orders of magnitude more computing resources than standard machine learning architectures.

Abstract:
To date, 14 high-redshift (z>1.0) galaxy clusters with mass measurements have been observed, spectroscopically confirmed and are reported in the literature. These objects should be exceedingly rare in the standard LCDM model. We conservatively approximate the selection functions of these clusters' parent surveys, and quantify the tension between the abundances of massive clusters as predicted by the standard LCDM model and the observed ones. We alleviate the tension considering non-Gaussian primordial perturbations of the local type, characterized by the parameter fnl and derive constraints on fnl arising from the mere existence of these clusters. At the 95% confidence level, fnl>467 with cosmological parameters fixed to their most likely WMAP5 values, or fnl > 123 (at 95% confidence) if we marginalize over WMAP5 parameters priors. In combination with fnl constraints from Cosmic Microwave Background and halo bias, this determination implies a scale-dependence of fnl at approx. 3 sigma. Given the assumptions made in the analysis, we expect any future improvements to the modeling of the non-Gaussian mass function, survey volumes, or selection functions to increase the significance of fnl>0 found here. In order to reconcile these massive, high-z clusters with an fnl=0, their masses would need to be systematically lowered by 1.5 sigma or the sigma8 parameter should be approx. 3 sigma higher than CMB (and large-scale structure) constraints. The existence of these objects is a puzzle: it either represents a challenge to the LCDM paradigme or it is an indication that the mass estimates of clusters is dramatically more uncertain than we think.

Abstract:
Using the publicly available VESPA database of SDSS Data Release 7 spectra, we calculate the stellar Mass Weighted Age (hereafter MWA) as a function of local galaxy density and dark matter halo mass. We compare our results with semi-analytic models from the public Millennium Simulation. We find that the stellar MWA has a large scatter which is inherent in the data and consistent with that seen in semi-analytic models. The stellar MWA is consistent with being independent (to first order) with local galaxy density, which is also seen in semi-analytic models. As a function of increasing dark matter halo mass (using the SDSS New York Value Added Group catalogues), we find that the average stellar MWA for member galaxies increases, which is again found in semi-analytic models. Furthermore we use public dark matter Mass Accretion History (MAH) code calibrated on simulations, to calculate the dark matter Mass Weighted Age as a function of dark matter halo mass. In agreement with earlier analyses, we find that the stellar MWA and the dark matter MWA are anti correlated for large mass halos, i.e, dark matter accretion does not seem to be the primary factor in determining when stellar mass was compiled. This effect can be described by down-sizing.

Abstract:
Supermassive black hole binary systems (SMBHB) are standard sirens -- the gravitational wave analogue of standard candles -- and if discovered by gravitational wave detectors, they could be used as precise distance indicators. Unfortunately, gravitational lensing will randomly magnify SMBHB signals, seriously degrading any distance measurements. Using a weak lensing map of the SMBHB line of sight, we can estimate its magnification and thereby remove some uncertainty in its distance, a procedure we call "delensing." We find that delensing is significantly improved when galaxy shears are combined with flexion measurements, which reduce small-scale noise in reconstructed magnification maps. Under a Gaussian approximation, we estimate that delensing with a 2D mosaic image from an Extremely Large Telescope (ELT) could reduce distance errors by about 30-40% for a SMBHB at z=2. Including an additional wide shear map from a space survey telescope could reduce distance errors by 50%. Such improvement would make SMBHBs considerably more valuable as cosmological distance probes or as a fully independent check on existing probes.

Abstract:
One of the most pressing issues in cosmology is whether general relativity (GR) plus a dark sector is the underlying physical theory or whether a modified gravity model is needed. Upcoming dark energy experiments designed to probe dark energy with multiple methods can address this question by comparing the results of the different methods in constraining dark energy parameters. Disagreement would signal the breakdown of the assumed model (GR plus dark energy). We study the power of this consistency test by projecting constraints in the $w_0-w_a$ plane from the four different techniques of the Dark Energy Survey in the event that the underlying true model is modified gravity. We find that the standard technique of looking for overlap has some shortcomings, and we propose an alternative, more powerful Multi-dimensional Consistency Test. We introduce the methodology for projecting whether a given experiment will be able to use this test to distinguish a modified gravity model from GR.

Abstract:
We critically investigate current statistical tests applied to high redshift clusters of galaxies in order to test the standard cosmological model and describe their range of validity. We carefully compare a sample of high-redshift, massive, galaxy clusters with realistic Poisson sample simulations of the theoretical mass function, which include the effect of Eddington bias. We compare the observations and simulations using the following statistical tests: the distributions of ensemble and individual existence probabilities (in the >M,>z sense), the redshift distributions, and the 2d Kolmogorov-Smirnov test. Using seemingly rare clusters from Hoyle et al. (2011), and Jee et al. (2011) and assuming the same survey geometry as in Jee et al. (2011, which is less conservative than Hoyle et al. 2011), we find that the (>M,>z) existence probabilities of all clusters are fully consistent with LCDM. However assuming the same survey geometry, we use the 2d K-S test probability to show that the observed clusters are not consistent with being the least probable clusters from simulations at >95% confidence, and are also not consistent with being a random selection of clusters, which may be caused by the non-trivial selection function and survey geometry. Tension can be removed if we examine only a X-ray selected sub sample, with simulations performed assuming a modified survey geometry.

Abstract:
We compare the stellar population properties in the central regions of visually classified non-starforming spiral and elliptical galaxies from Galaxy Zoo and SDSS DR7. The galaxies lie in the redshift range $0.04

Abstract:
We present an analysis of importance feature selection applied to photometric redshift estimation using the machine learning architecture Decision Trees with the ensemble learning routine Adaboost (hereafter RDF). We select a list of 85 easily measured (or derived) photometric quantities (or `features') and spectroscopic redshifts for almost two million galaxies from the Sloan Digital Sky Survey Data Release 10. After identifying which features have the most predictive power, we use standard artificial Neural Networks (aNN) to show that the addition of these features, in combination with the standard magnitudes and colours, improves the machine learning redshift estimate by 18% and decreases the catastrophic outlier rate by 32%. We further compare the redshift estimate using RDF with those from two different aNNs, and with photometric redshifts available from the SDSS. We find that the RDF requires orders of magnitude less computation time than the aNNs to obtain a machine learning redshift while reducing both the catastrophic outlier rate by up to 43%, and the redshift error by up to 25%. When compared to the SDSS photometric redshifts, the RDF machine learning redshifts both decreases the standard deviation of residuals scaled by 1/(1+z) by 36% from 0.066 to 0.041, and decreases the fraction of catastrophic outliers by 57% from 2.32% to 0.99%.

Abstract:
We present the clustering of galaxy clusters as a useful addition to the common set of cosmological observables. The clustering of clusters probes the large-scale structure of the Universe, extending galaxy clustering analysis to the high-peak, high-bias regime. Clustering of galaxy clusters complements the traditional cluster number counts and observable-mass relation analyses, significantly improving their constraining power by breaking existing calibration degeneracies. We use the maxBCG galaxy clusters catalogue to constrain cosmological parameters and cross-calibrate the mass-observable relation, using cluster abundances in richness bins and weak-lensing mass estimates. We then add the redshift-space power spectrum of the sample, including an effective modelling of the weakly non-linear contribution and allowing for an arbitrary photometric redshift smoothing. The inclusion of the power spectrum data allows for an improved self-calibration of the scaling relation. We find that the inclusion of the power spectrum typically brings a $\sim 50$ per cent improvement in the errors on the fluctuation amplitude $\sigma_8$ and the matter density $\Omega_{\mathrm{m}}$. Finally, we apply this method to constrain models of the early universe through the amount of primordial non-Gaussianity of the local type, using both the variation in the halo mass function and the variation in the cluster bias. We find a constraint on the amount of skewness $f_{\mathrm{NL}} = 12 \pm 157 $ ($1\sigma$) from the cluster data alone.

Abstract:
We examine the fraction of early-type (and spiral) galaxies found in groups and clusters of galaxies as a function of dark matter halo mass. We use morphological classifications from the Galaxy Zoo project matched to halo masses from both the C4 cluster catalogue and the Yang et al (2007) group catalogue. We find that the fraction of early-type (or spiral) galaxies remains constant (changing by less than 10%) over three orders of magnitude in halo mass (13