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
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Simulations find [C II] traces star formation robustly but underestimates outflow speeds and mass-loading factors by factors of 2-5, with feedback type affecting disk settling but not distinguishable from [C II] spatial or spectral properties alone.
Simulations from COSMOS2020 show masking recovers [CII] above 300 GHz in ideal conditions but noise prevents useful S/N until near the end of 2000-hour observations.
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.
-
Stellar feedback SPICEs up [C II] emission in the first galaxies
Simulations find [C II] traces star formation robustly but underestimates outflow speeds and mass-loading factors by factors of 2-5, with feedback type affecting disk settling but not distinguishable from [C II] spatial or spectral properties alone.
-
Testing masking effectiveness using multi-line image cubes based on COSMOS2020 for [CII] line intensity mapping at $z_{[CII]} > 3.5$
Simulations from COSMOS2020 show masking recovers [CII] above 300 GHz in ideal conditions but noise prevents useful S/N until near the end of 2000-hour observations.