Pith. sign in

REVIEW 1 cited by

Robustness of deep learning algorithms in astronomy -- galaxy morphology studies

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2111.00961 v2 pith:R43ANHQB submitted 2021-11-01 astro-ph.GA cs.CVcs.LG

classification astro-ph.GAcs.CVcs.LG
keywords modelsdatascientificadversarialcasecompressiondeepespecially
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Deep learning models are being increasingly adopted in wide array of scientific domains, especially to handle high-dimensionality and volume of the scientific data. However, these models tend to be brittle due to their complexity and overparametrization, especially to the inadvertent adversarial perturbations that can appear due to common image processing such as compression or blurring that are often seen with real scientific data. It is crucial to understand this brittleness and develop models robust to these adversarial perturbations. To this end, we study the effect of observational noise from the exposure time, as well as the worst case scenario of a one-pixel attack as a proxy for compression or telescope errors on performance of ResNet18 trained to distinguish between galaxies of different morphologies in LSST mock data. We also explore how domain adaptation techniques can help improve model robustness in case of this type of naturally occurring attacks and help scientists build more trustworthy and stable models.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Enhancing Galaxy Classification with U-Net Variational Autoencoders for Image Denoising

    astro-ph.IM 2025-06 reject novelty 5.0 of 10

    Applying a U-Net VAE denoising step to galaxy images before classification is reported to improve accuracy, reaching 97.45% with a GCNN on Galaxy10 DECaLS, although no direct noisy baseline is presented.

Pith tools