Agent swarm builds differentiable ice-flow PDE solver from real Antarctic satellite data
DynamicWebPaige · x · 2026-11-21
Dynamic Web Paige ran an agent swarm in Google Antigravity against real NASA MEaSUREs ITSLIVE satellite velocities (2015–2022) and BAS Bedmap3 bed topography over Thwaites and Pine Island Glaciers.
The agents built a differentiable 2D ice-flow PDE solver from scratch to map basal friction, achieving under 0.5% grounded-speed error and discovering sliding laws. Accompanying charts show Antarctic surface velocities, 7-year acceleration in the Amundsen Sea Embayment (Pine Island sped up 34.6 m/yr), plus effective pressure and bed roughness maps.
More from Research
- Ex-DeepMind researcher: if the brain is 'deep in time', are frontier models an abomination patched by looping? — yacinelearning · 2026-11-21
- Looped transformers: more looping boosts reasoning but with diminishing returns past 3 passes and fewer CoT tokens — nrehiew_ · 2026-11-21
- Team releases TML-d24-varloop-research on Hugging Face to study looping's effect on CoT monitorability — nrehiew_ · 2026-11-21
- Loop only the middle layers: design detail shows 1.51-1.91x compute-equivalent gain vs 1.24-1.36x for last layers — nrehiew_ · 2026-11-21
- New Company Agent Bench: Gemini Neon leads at 87.4%, simple API harnesses beat products — philhchen · 2026-11-21
- NEJM Analysis of $45B in Medicare Payments: Pathology 99% AI-Exposed, Dermatology 7% — HealthcareAIGuy · 2026-11-21