AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Practical Global and Local Bounds in Gaussian Process Regression via Chaining

Junyi Liu, Stanley Kok

AAAI 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: PAC-Bayes and decoupled GP learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Matern-specific chaining bound instead of RBF-specific bound (ablation)chaining bounds · Loa_CO2, pointwise 99%CWC: 0.81 -> 0.24Improved
use Matern-specific chaining bound instead of RBF-specific bound (ablation)chaining bounds · Boston, pointwise 99%CWC: 1.01 -> 0.76Improved
use Matern-specific chaining bound instead of RBF-specific bound (ablation)chaining bounds · Sarcos, expected upper/lower boundsCWC: 1.03 -> 0.78Improved
use Matern-specific chaining bound instead of RBF-specific bound (ablation)chaining bounds · Loa_CO2, expected upper/lower boundsCWC: 1.70 -> 2.08Hurt
use Matern-specific chaining bound instead of RBF-specific bound (ablation)chaining bounds · Auto-mpg, expected upper/lower boundsCWC: 3.06 -> 3.24Hurt
use chaining bounds (Matern) instead of Capone22Gaussian process bounds · Loa_CO2, pointwise 99%CWC: 1761.85 -> 0.24Improved
use chaining bounds (Matern) instead of Lederer19Gaussian process bounds · Auto-mpg, pointwise 99%CWC: 50.03 -> 0.84Improved
use chaining bounds (RBF) instead of Bayesian CIGaussian process bounds · Synthetic, pointwise 99%CWC: 37.19 -> 1.80Improved
use chaining bounds (Matern) instead of Fiedler21Gaussian process bounds · Loa_CO2, expected upper/lower boundsCWC: 16.01 -> 2.08Improved
use chaining bounds (RBF) instead of Fiedler21Gaussian process bounds · Synthetic, pointwise 95%CWC: 1.49 -> 1.50No clear effect
use chaining bounds instead of Capone22Gaussian process bounds · Auto-mpg, expected upper/lower boundsCWC: 2.81 -> 3.06Hurt
use chaining bounds (RBF) instead of Lederer19Gaussian process bounds · Loa_CO2, expected upper/lower boundsCWC: 1.73 -> 1.70Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy