Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Efficiently sampling functions from Gaussian process posteriors
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 decoupled sampling instead of iterative conditioning | Sparse GP simulation of FitzHugh-Nagumo dynamics · FitzHugh-Nagumo simulation | runtime: 10 hours -> 20 seconds | Helped |
| Use decoupled sampling rather than weight-space sampling | GP posterior sampling · 2-Wasserstein sample tests | 2-Wasserstein distance | Helped |
| Use decoupled sampling rather than function-space TS, weight-space TS, or the stated baselines | Parallel Thompson sampling · GP-prior optimization tasks | optimization performance | Helped |
| Use LOVE sampling instead of decoupled sampling | Exact GP posterior sampling · 2-Wasserstein sample tests | 2-Wasserstein distance | Hurt |
| Use a finite Fourier-feature approximation without a function-space update | GP posterior sampling · Increasing training data relative to basis count | posterior sample quality | Hurt |
| Represent the sparse posterior entirely in weight-space | Sparse GP posterior sampling · Alternative decomposition | posterior sample quality | Hurt |
| Use a Fourier-feature approximation for the orthogonal residual component | Orthogonal decomposition of sparse GP posterior · Alternative decomposition | posterior sample quality | Hurt |
Closest papers by meaning
- Turbocharging Gaussian Process Inference with Approximate Sketch-and-Project (NeurIPS 2025 · similarity 0.91)
- Conditioning Gaussian Processes on Almost Anything (NeurIPS 2026 · similarity 0.90)
- Sampling from Gaussian Process Posteriors using Stochastic Gradient Descent (NeurIPS 2023 · similarity 0.90)
- Computation-Aware Gaussian Processes: Model Selection And Linear-Time Inference (NeurIPS 2024 · similarity 0.88)
- Bayesian Optimization through Gaussian Cox Process Models for Spatio-temporal Data (ICLR 2024 · similarity 0.87)
- Robust and Computation-Aware Gaussian Processes (NeurIPS 2025 · similarity 0.87)
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.