{"slug": "learning-calibrated-and-samplable-predictive-distributions-of-surface-ocean-p-co", "title": "Learning Calibrated and Samplable Predictive Distributions of Surface-Ocean $\\Delta p$CO$_2$ Using Tabular Diffusion Models", "summary": "A conditional diffusion model called Treeffuser, using gradient-boosted-tree score functions, produced the best-calibrated predictive distributions of sea-air pCO2 difference (ΔpCO2) among the uncertainty-quantification methods evaluated on a held-out SOCAT test set, according to the arXiv paper 2611.21769v1. Treeffuser achieved the lowest continuous ranked probability score at 8.229 μatm and coverage closest to nominal at 0.686 for 68% and 0.930 for 95%, while its point-prediction skill stayed comparable to a deterministic XGBoost baseline (RMSE 18.4 vs. 17.7 μatm; R² 0.77 vs. 0.79). The authors state the results show conditional diffusion can supply well-calibrated, observation-level uncertainties while largely preserving point-prediction skill, and that spatially coherent field reconstruction will require modeling dependence across grid cells and evaluation on spatially held-out data.", "body_md": "arXiv:2611.21769v1 Announce Type: new \nAbstract: The ocean absorbs almost 30% of anthropogenic CO$_2$ emissions, but quantifying this sink requires estimating surface-ocean CO$_2$ from observations that are sparse in space and time. Most existing machine-learning estimates of this field are deterministic, giving a single value per grid point, with uncertainty approximated post hoc via residuals, ensembles, or prescribed distributions. We instead learn a conditional predictive distribution for each observation: using Treeffuser, a conditional diffusion model with gradient-boosted-tree score functions, we estimate the full predictive distribution of $\\Delta p$CO$_2$ (the sea-air $p$CO$_2$ difference) from 13 environmental, spatial, and temporal predictors. On a held-out SOCAT test set, Treeffuser attains the lowest continuous ranked probability score among the uncertainty-quantification methods evaluated (8.229 $\\mu$atm) and the coverage closest to the nominal 68% / 95% levels (0.686 / 0.930), while retaining point-prediction skill comparable to a deterministic XGBoost baseline (RMSE 18.4 vs. 17.7 $\\mu$atm; $R^2$ 0.77 vs. 0.79). These results show that conditional diffusion can provide well-calibrated, observation-level uncertainties while largely preserving point-prediction skill. Because Treeffuser is generative, its samples can support probabilistic analyses of $\\Delta p$CO$_2$; extending this to spatially coherent field reconstruction will require modeling dependence across grid cells and evaluation on spatially held-out data.", "url": "https://wpnews.pro/news/learning-calibrated-and-samplable-predictive-distributions-of-surface-ocean-p-co", "canonical_source": "https://arxiv.org/abs/2611.21769", "published_at": "2026-11-18 05:00:00+00:00", "updated_at": "2026-11-18 05:16:59.499755+00:00", "lang": "en", "topics": ["machine-learning", "ai-research", "neural-networks"], "entities": ["Treeffuser", "SOCAT", "XGBoost", "arXiv"], "also_reported_by": [], "takeaway": "Treeffuser, a conditional diffusion model with gradient-boosted-tree score functions, reached a CRPS of 8.229 μatm and coverage of 0.686/0.930 against nominal 68%/95% on held-out SOCAT data, versus an XGBoost baseline's RMSE of 17.7 μatm and R² of 0.79.", "recommendation": null, "alternates": {"html": "https://wpnews.pro/news/learning-calibrated-and-samplable-predictive-distributions-of-surface-ocean-p-co", "markdown": "https://wpnews.pro/news/learning-calibrated-and-samplable-predictive-distributions-of-surface-ocean-p-co.md", "text": "https://wpnews.pro/news/learning-calibrated-and-samplable-predictive-distributions-of-surface-ocean-p-co.txt", "jsonld": "https://wpnews.pro/news/learning-calibrated-and-samplable-predictive-distributions-of-surface-ocean-p-co.jsonld"}}