Photometric Redshifts for Cosmology: Improving Accuracy and Uncertainty Estimates Using Bayesian Neural Networks

Bernie Boscoe; Evan Jones; Jack Singal; Tuan Do; Yujie Wan; Zooey Nguyen

RECORD

Photometric Redshifts for Cosmology: Improving Accuracy and Uncertainty Estimates Using Bayesian Neural Networks

Evan Jones, Tuan Do, Bernie Boscoe, Yujie Wan, Zooey Nguyen, Jack Singal

Cosmology and Nongalactic Astrophysics · Submitted 15 Feb 2022

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim We find that while the Bayesian neural network did not perform as well as non-Bayesian neural networks if evaluated solely by point estimate photo-z values, BNNs can provide uncertainty estimates that are necessary for cosmology

how it is stated · asserted 0.97 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2202.07121 (version 1)

Pith IDGCNDPPS2, Pith's permanent record of this paper and its review

CitationBernie Boscoe, Evan Jones, Jack Singal, Tuan Do, Yujie Wan, Zooey Nguyen. "Photometric Redshifts for Cosmology: Improving Accuracy and Uncertainty Estimates Using Bayesian Neural Networks." arXiv:2202.07121 (2022). https://pith.science/paper/GCNDPPS2

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 5,724 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.93 · human participants 0.02 · controlled comparison 0.75 · randomized assignment 0.01 · states quantitative result 0.06 · compares with prior results 0.89 · mentions code or data availability 0.04.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

We present results exploring the role that probabilistic deep learning models can play in cosmology from large scale astronomical surveys through estimating the distances to galaxies (redshifts) from photometry. Due to the massive scale of data coming from these new and upcoming sky surveys, machine learning techniques using galaxy photometry are increasingly adopted to predict galactic redshifts which are important for inferring cosmological parameters such as the nature of dark energy. Associated uncertainty estimates are also critical measurements, however, common machine learning methods typically provide only point estimates and lack uncertainty information as outputs. We turn to Bayesian neural networks (BNNs) as a promising way to provide accurate predictions of redshift values. We have compiled a new galaxy training dataset from the Hyper Suprime-Cam Survey, designed to mimic large surveys, but over a smaller portion of the sky. We evaluate the performance and accuracy of photometric redshift (photo-z) predictions from photometry using machine learning, astronomical and probabilistic metrics. We find that while the Bayesian neural network did not perform as well as non-Bayesian neural networks if evaluated solely by point estimate photo-z values, BNNs can provide uncertainty estimates that are necessary for cosmology

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	We present results exploring the role that probabilistic deep learning models can play in cosmology from large scale astronomical surveys through estimating the distances to galaxies (redshifts) from photometry.	contribution 0.98	0.84	0.06	asserted 0.77
s1	Due to the massive scale of data coming from these new and upcoming sky surveys, machine learning techniques using galaxy photometry are increasingly adopted to predict galactic redshifts which are important for inferring cosmological parameters such as the nature of dark energy.	background 1.00	0.06	0.03	asserted 0.99
s2	Associated uncertainty estimates are also critical measurements, however, common machine learning methods typically provide only point estimates and lack uncertainty information as outputs.	gap 0.89	0.08	0.04	asserted 1.00
s3	We turn to Bayesian neural networks (BNNs) as a promising way to provide accurate predictions of redshift values.	method 0.79	0.41	0.07	hedged possibility 0.82
s4	We have compiled a new galaxy training dataset from the Hyper Suprime-Cam Survey, designed to mimic large surveys, but over a smaller portion of the sky.	method 0.83	0.96	0.02	asserted 1.00
s5	We evaluate the performance and accuracy of photometric redshift (photo-z) predictions from photometry using machine learning, astronomical and probabilistic metrics.	method 0.86	0.92	0.03	asserted 0.83
s6	We find that while the Bayesian neural network did not perform as well as non-Bayesian neural networks if evaluated solely by point estimate photo-z values, BNNs can provide uncertainty estimates that are necessary for cosmology	result 0.91	0.96	0.85	asserted 0.97

What kind of paper the classifier reports this is: empirical study 0.48 · evidence from analysis of existing data 0.68.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

We present results exploring the role that probabilistic deep learning models can play in cosmology from large scale astronomical surveys through estimating the distances to galaxies (redshifts) from photometry. Due to the massive scale of data coming from these new and upcoming sky surveys, machine learning techniques using galaxy photometry are increasingly adopted to predict galactic redshifts which are important for inferring cosmological parameters such as the nature of dark energy. Associated uncertainty estimates are also critical measurements, however, common machine learning methods typically provide only point estimates and lack uncertainty information as outputs. We turn to Bayesian neural networks (BNNs) as a promising way to provide accurate predictions of redshift values. We have compiled a new galaxy training dataset from the Hyper Suprime-Cam Survey, designed to mimic large surveys, but over a smaller portion of the sky. We evaluate the performance and accuracy of photometric redshift (photo-z) predictions from photometry using machine learning, astronomical and probabilistic metrics. We find that while the Bayesian neural network did not perform as well as non-Bayesian neural networks if evaluated solely by point estimate photo-z values, BNNs can provide uncertainty estimates that are necessary for cosmology

Human review

How Pith checks papers → · Full audit of this paper →

References

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

AquiLLM: Evaluating Faithfulness in Open-Weight RAG-LLM Systems for Scientific Research
cs.AI 2026-09 conditional novelty 5.0 of 10

AquiLLM, an open-weight RAG-LLM system, shows strong faithfulness on retrieval tasks but degrades on synthesis and comparative analysis in an astronomy case study.
Using different sources of ground truths and transfer learning to improve the generalization of photometric redshift estimation
astro-ph.IM 2024-11 conditional novelty 5.0 of 10

Transfer learning from a broad photometric-redshift sample to a spectroscopic sample cuts bias and RMS error for galaxy redshift prediction on the spectroscopic sample, but degrades performance on the broad sample.

How the literature uses it

5 arXiv papers cite this in 6 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it · 2 sentences without a confident label

Uses its method or tool 2 papers

We employed a neural network (NN) architecture based on [ref] for photometric redshift estimation, consisting of four fully connected layers with ReLU activation and a skip connection.

Using different sources of ground truths and transfer learning to improve the generalization of photometric redshift estimation · Methodology & metrics · 1.00
And we define a similar metric called coverage introduced in [ref] to examine our reliability of uncertainties: [equation] where $N_gal$ is the number of galaxy samples in specific redshift bin, and $σ$ is the 68% confidence interval of prediction for every sample.

Photometric redshift estimates using Bayesian neural networks in the CSST survey · Results and Discussions · 0.92

Cites it as background 1 paper

Photo-$z$ estimation techniques fall into two categories: template-fitting [ref] , which matches photometry to spectroscopic or physical model templates, and machine learning [ref] , which trains supervised models to map photometry to reference redshifts.

Zephyr : Stitching Heterogeneous Training Data with Normalizing Flows for Photometric Redshift Inference · Introduction · 0.92

What it cites · 2 sentences without a confident label

Uses the method or tool of 1 paper

We use data quality cuts similar to [ref] and [ref] (see [<https://doi.org/10.5281/zenodo.5528827>] for full list).

Photometric Redshifts for the Hyper Suprime-Cam Subaru Strategic Program Data Release 2 · Data: Galaxy observations · 0.94

Uses data or values from 1 paper

We crossmatched galaxy photometry from HSC with the HSC collection of publicly available spectroscopic redshifts [ref] , [ref] , [ref] , [ref] , [ref] , [ref] , [ref] , [ref] using the galaxies' sky positions (d < 1 arcsecond).

Science Objectives and Early Results of the DEEP2 Redshift Survey · Data: Galaxy observations · 0.99

Compares with 1 paper

While our goal in this work is to compare the two types of NNs, we provide a comparison to LSST requirements [ref] (Table 2) for reference.

The LSST Dark Energy Science Collaboration (DESC) Science Requirements Document · Results · 0.94

Cites as background 1 paper

In this work, we investigate photometric redshift (photo-z) estimation using Bayesian neural networks (BNN), a type of probabilistic neural network (NN) [ref] .

Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users · Introduction · 0.92

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.