Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Leveraging Locality and Robustness to Achieve Massively Scalable Gaussian Process Regression
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use GPnn instead of SVGP | GP regression · Poletele | RMSE: 0.226 ± 0.0059 -> 0.195 ± 0.0042 | Helped |
| Use GPnn instead of SVGP | GP regression · Bike | RMSE: 0.606 ± 0.0033 -> 0.624 ± 0.0079 | Hurt |
| Use GPnn instead of SVGP | GP regression · Protein | RMSE: 0.688 ± 0.0043 -> 0.666 ± 0.0014 | Helped |
| Use GPnn instead of SVGP | GP regression · Ctslice | RMSE: 0.384 ± 0.0064 -> 0.132 ± 0.00062 | Helped |
| Use GPnn instead of SVGP | GP regression · Road3D | RMSE: 0.443 ± 0.008 -> 0.351 ± 0.0014 | Helped |
| Use GPnn instead of SVGP | GP regression · Song | RMSE: 0.834 ± 0.0011 -> 0.787 ± 0.0045 | Helped |
| Use GPnn instead of SVGP | GP regression · HouseE | RMSE: 0.0566 ± 0.00011 -> 0.0506 ± 0.00072 | Helped |
| Calibrate the noise and kernel-scale parameters using held-out training examples | GPnn · GPnn calibration set | weak calibration | Helped |
| Calibrate the noise and kernel-scale parameters using held-out training examples | GPnn · GPnn calibration set | NLL | Helped |
| Calibrate the noise and kernel-scale parameters using held-out training examples | GPnn · Any test set | MSE | No clear effect |
| Replace RBF with the exponential kernel | GPnn · Road3D | RMSE: 0.351 to 0.098 | Helped |
| Set the noise-variance estimate to the true noise variance | GPnn · Large-n predictions under misspecification | NLL | Helped |
Closest papers by meaning
- Robust and Computation-Aware Gaussian Processes (NeurIPS 2025 · similarity 0.90)
- Computation-Aware Gaussian Processes: Model Selection And Linear-Time Inference (NeurIPS 2024 · similarity 0.89)
- Error Bounds for Gaussian Process Regression Under Bounded Support Noise with Applications to Safety Certification (AAAI 2025 · similarity 0.89)
- Robust and Conjugate Gaussian Process Regression (ICML 2024 · similarity 0.88)
- Robust Gaussian Processes via Relevance Pursuit (NeurIPS 2024 · similarity 0.87)
- New Bounds for Sparse Variational Gaussian Processes (ICML 2025 · similarity 0.86)
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.