HealDA: Highlighting the importance of initial errors in end-to-end AI weather forecasts

Aayush Gupta, Akshay Subramaniam, Michael S. Pritchard, Karthik Kashinath, Sergey Frolov, Kelsey Lieberman, Christopher Miller, Nicholas Silverman, Noah D. Brenowitz · 2026

Read HealDA: Highlighting the importance of initial errors in end-to-end AI weather forecasts by Aayush Gupta, Akshay Subramaniam, Michael S. Pritchard, Karthik Kashinath, Sergey Frolov, Kelsey Lieberman, Christopher Miller, Nicholas Silverman, Noah D. Brenowitz (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-17636.txt.

Description

Aayush Gupta and five co-authors present HealDA, a fix for a weak point in AI weather forecasts. Machine learning weather models now rival top classic systems at medium range. Almost all still need classic data assimilation for start conditions. That tie keeps them bound to costly infrastructure. It also caps their speed and accuracy gains. HealDA stresses the role of initial errors.

Initial errors grow through a forecast and set its fate. Highlighting them can guide better assimilation for ML models. A free-standing ML pipeline would run faster and cheaper. That could widen access to top-grade forecasts. The work targets the handoff where ML still leans on the old system.

Catalog record

CREATORAayush Gupta, Akshay Subramaniam, Michael S. Pritchard, Karthik Kashinath, Sergey Frolov, Kelsey Lieberman, Christopher Miller, Nicholas Silverman, Noah D. Brenowitz
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE65 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-17636

Browse the full library · About the Tower of Babel