arXiv subjects

Irene Moskowitz

Publications and source records attributed to Irene Moskowitz.

5 recordsLinked to original sources

arXiv2026

Diagnosing the Effects of Spectroscopic Training Set Imperfection on Photometric Redshift Performance

Most LSST extragalactic science will rely on photometric redshifts (photo-$z$) to extract distance information for the galaxies. However, an incomplete or non-representative training set can introduce bias into photo-$z$ estimation. It is necessary to understand how various forms of training set imperfection, such as incompleteness and non-trivial spectroscopic target selection, affect photo-$z$ estimation algorithms, and to identify metrics best-suited to quantify the impact. This work aims to systematically study metrics for diagnosing how various photo-$z$ methods react to certain types of training set incompleteness and non-representativeness. We use methods available through the open-source Python library Redshift Assessment Infrastructure Layers (RAIL) to systematically test the algorithms CMNN, GPz, FlexZBoost, and PZFlow on mock training data degraded in accordance with several existing spectroscopic sky surveys, as well as under conditions of inverse redshift incompleteness, which approximately mimics observed patterns of incompleteness at high redshift. We employ the algorithm TrainZ as a control. Finally, we quantify photo-$z$ algorithm performance using a variety of statistical metrics implemented externally to RAIL. We determine that the Kullback-Liebler Divergence, Wasserstein Distance, and Probability Integral Transform are particularly informative metrics with which to assess the impact of training set imperfection on algorithmic performance. We also find that inverse redshift incompleteness effects alone lack the complexity to realistically represent anticipated training data.

astro-ph.IM↗

arXiv2025

Improved photometric redshift estimations through self-organising map-based data augmentation

We introduce a framework for the enhanced estimation of photometric redshifts using Self-Organising Maps (SOMs). Our method projects galaxy Spectral Energy Distributions (SEDs) onto a two-dimensional map, identifying regions that are sparsely sampled by existing spectroscopic observations. These under-sampled areas are then augmented with simulated galaxies, yielding a more representative spectroscopic training dataset. To assess the efficacy of this SOM-based data augmentation in the context of the forthcoming Legacy Survey of Space and Time (LSST), we employ mock galaxy catalogues from the OpenUniverse2024 project and generate synthetic datasets that mimic the expected photometric selections of LSST after one (Y1) and ten (Y10) years of observation. We construct 501 degraded realisations by sampling galaxy colours, magnitudes, redshifts and spectroscopic success rates, in order to emulate the compilation of a wide array of realistic spectroscopic surveys. Augmenting the degraded mock datasets with simulated galaxies from the independent CosmoDC2 catalogues has markedly improved the performance of our photometric redshift estimates compared to models lacking this augmentation, particularly for high-redshift galaxies ($z_\mathrm{true} \gtrsim 1.5$). This improvement is manifested in notably reduced systematic biases and a decrease in catastrophic failures by up to approximately a factor of 2, along with a reduction in information loss in the conditional density estimations. These results underscore the effectiveness of SOM-based augmentation in refining photometric redshift estimation, thereby enabling more robust analyses in cosmology and astrophysics for the NSF-DOE Vera C. Rubin Observatory.

astro-ph.GA↗

arXiv2025

Redshift Assessment Infrastructure Layers (RAIL): Rubin-era photometric redshift stress-testing and at-scale production

Virtually all extragalactic use cases of the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) require the use of galaxy redshift information, yet the vast majority of its sample of tens of billions of galaxies will lack high-fidelity spectroscopic measurements thereof, instead relying on photometric redshifts (photo-$z$) subject to systematic imprecision and inaccuracy best encapsulated by photo-$z$ probability density functions (PDFs). We present the version 1 release of Redshift Assessment Infrastructure Layers (RAIL), an open source Python library for at-scale probabilistic photo-$z$ estimation, initiated by the LSST Dark Energy Science Collaboration (DESC) with contributions from the LSST Interdisciplinary Network for Collaboration and Computing (LINCC) Frameworks team. RAIL's three subpackages provide modular tools for end-to-end stress-testing, including a forward modeling suite to generate realistically complex photometry, a unified API for estimating per-galaxy and ensemble redshift PDFs by an extensible set of algorithms, and built-in metrics of both photo-$z$ PDFs and point estimates. RAIL serves as a flexible toolkit enabling the derivation and optimization of photo-$z$ data products at scale for a variety of science goals and is not specific to LSST data. We thus describe to the extragalactic science community, including and beyond Rubin the design and functionality of the RAIL software library so that any researcher may have access to its wide array of photo-$z$ characterization and assessment tools.

astro-ph.IM↗

arXiv2024

Improving Photometric Redshift Estimates with Training Sample Augmentation

Large imaging surveys will rely on photometric redshifts (photo-z's), which are typically estimated through machine learning methods. Currently planned spectroscopic surveys will not be deep enough to produce a representative training sample for LSST, so we seek methods to improve the photo-z estimates that arise from non-representative training samples. Spectroscopic training samples for photo-z's are biased towards redder, brighter galaxies, which also tend to be at lower redshift than the typical galaxy observed by LSST, leading to poor photo-z estimates with outlier fractions nearly 4 times larger than for a representative training sample. In this paper, we apply the concept of training sample augmentation, where we augment simulated non-representative training samples with simulated galaxies possessing otherwise unrepresented features. When we select simulated galaxies with (g-z) color, i-band magnitude and redshift outside the range of the original training sample, we are able to reduce the outlier fraction of the photo-z estimates for simulated LSST data by nearly 50% and the normalized median absolute deviation (NMAD) by 56%. When compared to a fully representative training sample, augmentation can recover nearly 70% of the degradation in the outlier fraction and 80% of the degradation in NMAD. Training sample augmentation is a simple and effective way to improve training samples for photo-z's without requiring additional spectroscopic samples.

astro-ph.IM↗

arXiv2022

Improved Tomographic Binning of 3x2pt Lens Samples: Neural Network Classifiers and Optimal Bin Assignments

Large imaging surveys, such as the Legacy Survey of Space and Time, rely on photometric redshifts and tomographic binning for 3x2pt analyses that combine galaxy clustering and weak lensing. In this paper, we propose a method for optimizing the tomographic binning choice for the lens sample of galaxies. We divide the CosmoDC2 and Buzzard simulated galaxy catalogs into a training set and an application set, where the training set is nonrepresentative in a realistic way, and then estimate photometric redshifts for the application sets. The galaxies are sorted into redshift bins covering equal intervals of redshift or comoving distance, or with an equal number of galaxies in each bin, and we consider a generalized extension of these approaches. We find that bins of equal comoving distance produce the highest dark energy figure of merit of the initial binning choices, but that the choice of bin edges can be further optimized. We then train a neural network classifier to identify galaxies that are either highly likely to have accurate photometric redshift estimates or highly likely to be sorted into the correct redshift bin. The neural network classifier is used to remove poor redshift estimates from the sample, and the results are compared to the case when none of the sample is removed. We find that the neural network classifiers are able to improve the figure of merit by ~13% and are able to recover ~25% of the loss in the figure of merit that occurs when a nonrepresentative training sample is used.

astro-ph.CO↗