PITA, a new semi-supervised deep learning algorithm, outperforms prior photo-z methods by using a triple-task loss on images, colors, and available redshifts to produce a smooth latent space.
Title resolution pending
7 Pith papers cite this work. Polarity classification is still indexing.
representative citing papers
Euclid identifies 16 massive compact galaxies with V-shaped SEDs at z>4, half as old as the universe at their redshift, mostly distinct from known AGN.
PASSAGE releases 2183 spectroscopic redshifts (0.08<z<4.7) from JWST NIRISS observations in COSMOS, with strong agreement to photometric redshifts for multi-line sources and a quantified ~18% misidentification rate for single-line emitters.
Observational analysis of COSMOS galaxy groups shows quiescent fraction builds steadily from z~1.5-2 with faster growth in richer groups, first RS ridgeline galaxies by z~2, a possible early overdensity at z=3.4, and no significant evolution in RS slope or scatter.
Cross-correlation of CLAMATO Lyman-alpha forest with COSMOS galaxies yields stellar-mass-dependent biases of approximately 2.1, 3.2, and 3.8, corresponding to halo masses of log M_h ~ 10.5, 11.7, and 12.1 from Bolshoi-Planck mocks, with hints of enhanced low-mass star formation.
The N2CLS 2mm survey measures a mean S2/S1.2 color of 0.222 and mean redshifts of 3.2-3.6, finds no exotic 2mm-only population, and shows 2mm samples miss z~2 galaxies compared to 1.2mm samples.
The NIKA2 survey delivers catalogs of 323 mm-selected sources in COSMOS with redshifts peaking at z=2.8, including 66 at z>4, matching SIDES simulations but inconsistent with four other galaxy evolution models.
citing papers explorer
-
Optimizing Deep Learning Photometric Redshifts for the Roman Space Telescope with HST/CANDELS
PITA, a new semi-supervised deep learning algorithm, outperforms prior photo-z methods by using a triple-task loss on images, colors, and available redshifts to produce a smooth latent space.
-
Euclid: Scaled-up little red dots and other sources with v-shaped spectral energy distributions at z>4
Euclid identifies 16 massive compact galaxies with V-shaped SEDs at z>4, half as old as the universe at their redshift, mostly distinct from known AGN.
-
The NIRISS PASSAGE Spectroscopic Redshift Catalog in COSMOS
PASSAGE releases 2183 spectroscopic redshifts (0.08<z<4.7) from JWST NIRISS observations in COSMOS, with strong agreement to photometric redshifts for multi-line sources and a quantified ~18% misidentification rate for single-line emitters.
-
COSMOS-Web galaxy groups: Evolution of red sequence and quiescent galaxy fraction
Observational analysis of COSMOS galaxy groups shows quiescent fraction builds steadily from z~1.5-2 with faster growth in richer groups, first RS ridgeline galaxies by z~2, a possible early overdensity at z=3.4, and no significant evolution in RS slope or scatter.
-
Cross-correlations between the CLAMATO Lyman-alpha forest and galaxies within the COSMOS field
Cross-correlation of CLAMATO Lyman-alpha forest with COSMOS galaxies yields stellar-mass-dependent biases of approximately 2.1, 3.2, and 3.8, corresponding to halo masses of log M_h ~ 10.5, 11.7, and 12.1 from Bolshoi-Planck mocks, with hints of enhanced low-mass star formation.
-
The NIKA2 cosmological legacy survey at 2 mm: catalogs, colors, redshift distributions, and implications for deep surveys
The N2CLS 2mm survey measures a mean S2/S1.2 color of 0.222 and mean redshifts of 3.2-3.6, finds no exotic 2mm-only population, and shows 2mm samples miss z~2 galaxies compared to 1.2mm samples.
-
The NIKA2 Cosmological Legacy Survey in COSMOS: Final 1.2mm and 2mm source catalogs and redshift distribution of dusty star-forming galaxies
The NIKA2 survey delivers catalogs of 323 mm-selected sources in COSMOS with redshifts peaking at z=2.8, including 66 at z>4, matching SIDES simulations but inconsistent with four other galaxy evolution models.