Antigravity subsurface imaging: 16 posterior slices, 32% higher uncertainty flagged in shadow zones
DynamicWebPaige · x · 2026-11-21
DynamicWebPaige walks through Bayesian uncertainty quantification for subsurface imaging using the Antigravity system.
- Instead of one guess, the system builds 16 plausible underground velocity slices (RTO-SVGD ensemble, K=16) and automatically flags 32% higher uncertainty in the dark 'shadow zone' under the salt, with 90% error bars.
- The data also shows where the tool struggles: -54% error on salt domes, -34% on earthquake-style faults, best overall on buried CO₂ tracking — but plain sound waves win on an extreme broken-gravity test.
- Charts cover posterior mean, per-point std, true-error correspondence, credible profiles, and calibration curves, with under-coverage in the shadow zone explicitly exposed.
More from Research
- Alessandro Masullo announces a purely human proof of the Riemann hypothesis — davilagrau · 2026-11-21
- LLM chain-of-thought hurts Q-matrix accuracy: 80.9% vs 87.7% expert match, study finds — alex_verem · 2026-11-21
- Once AI proves it, shrinking a 250,000-page proof becomes the interesting problem — ZeroStateReflex · 2026-11-21
- 'Most algorithmic progress is data progress': why data quality beats clever algorithms — teortaxesTex · 2026-11-21
- JapanFold on Tenstorrent silicon predicts 1M+ protein structures powering Anthropic design contest — DavidBennett__ · 2026-11-21
- Open Amazon deforestation prediction leaderboard launches, top model hits 74% — anselm · 2026-11-21