Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Variational nearest neighbor Gaussian process
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 |
|---|---|---|---|
| replace SVGP low-rank prior with VNNGP nearest-neighbor sparse precision prior (inducing points at all observed locations) | variational GP on large datasets · Table 1 NLL across 11 datasets | test NLL | Mixed |
| VNNGP nearest-neighbor prior vs SVGP 1024 inducing points | spatial datasets (Precipitation, UKHousing, Covtype) · Precipitation, UKHousing, Covtype | test NLL | Helped |
| increase number of nearest neighbors K (32 to 256) (ablation) | VNNGP · Elevators and UKHousing | test RMSE and NLL | Helped |
| VNNGP uses the nearest neighbor KL (factorized over inducing points) instead of SWSGP local KL (ablation) | VNNGP vs SWSGP on high-noise simulation · simulated noisy GP posterior | posterior quality in data-sparse region | Helped |
| SWSGP with inducing points at all observed locations (SWSGP-N) instead of SWSGP-M (ablation) | SWSGP · Protein, 3DRoad, PoleTele | test NLL | Hurt |
| ELBO-based model selection of likelihood noise: VNNGP vs SWSGP with M=N (ablation) | VNNGP vs SWSGP-N · UKHousing noise selection | noise selected / test NLL | Helped |
Closest papers by meaning
- Sparse within Sparse Gaussian Processes using Neighbor Information (ICML 2021 · similarity 0.91)
- Input Dependent Sparse Gaussian Processes (ICML 2022 · similarity 0.91)
- Sparse Gaussian Processes: Structured Approximations and Power-EP Revisited (NeurIPS 2025 · similarity 0.90)
- Variational Gaussian Processes with Decoupled Conditionals (NeurIPS 2023 · similarity 0.90)
- New Bounds for Sparse Variational Gaussian Processes (ICML 2025 · similarity 0.89)
- Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling (ICML 2025 · similarity 0.88)
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.