Research map / Probabilistic and scientific modeling / Variational and generative sampling
Intrinsic Gaussian Process on Unknown Manifolds with Probabilistic Metrics
mu niu, Zhenwen Dai, Pokman Cheung, Yizhu Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use heat kernel on learned manifold (GPUM) instead of Euclidean R3 GP | GP regression · Swiss roll v=450 | mean RMSE: 0.298 -> 0.162 | Improved |
| Use GPUM instead of latent-space Euclidean R2 GP | GP regression · Swiss roll v=450 | mean RMSE: 0.290 -> 0.162 | Improved |
| Use GPUM instead of graph Laplacian GL-GP | GP regression · Swiss roll v=250 | mean RMSE: 0.243 -> 0.163 | Improved |
| Use GPUM instead of graph matern GM-GP | GP regression · Swiss roll v=800 | mean RMSE: 0.206 -> 0.164 | Improved |
| Increase unlabeled grid points from 250 to 800 | GL-GP · Swiss roll | mean RMSE: 0.243 -> 0.216 | Improved |
| Increase unlabeled grid points from 250 to 800 | GPUM · Swiss roll | mean RMSE: 0.163 -> 0.164 | No clear effect |
| Use GPUM instead of R2 GP | GP regression · WiFi location (n=3 labeled) | mean RMSE: 4.83 -> 4.11 | Improved |
| Use GPUM instead of graph-based GL-GP and GM-GP with few points | GP regression · WiFi location (n+v=36) | mean RMSE: 5.6 -> 4.11 | Improved |
| Use GPUM instead of R1024 GP | GP regression · COIL camera angle | mean RMSE: 0.097 -> 0.060 | No clear effect |
| Use GPUM instead of GL-GP and GM-GP | GP regression · COIL camera angle | mean RMSE: 0.116 -> 0.060 | Improved |
Closest papers by meaning
- Inference for Gaussian Processes with Matern Covariogram on Compact Riemannian Manifolds (NeurIPS 2023 · similarity 0.90)
- Posterior Contraction Rates for Matérn Gaussian Processes on Riemannian Manifolds (NeurIPS 2023 · similarity 0.87)
- Residual Deep Gaussian Processes on Manifolds (ICLR 2025 · similarity 0.87)
- Implicit Gaussian process representation of vector fields over arbitrary latent manifolds (ICLR 2024 · similarity 0.85)
- Intrinsic Riemannian Cross-covariance for Manifold-valued Random Objects (NeurIPS 2026 · similarity 0.83)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.82)
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.
