Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Gaussian Processes with Errors in Variables: Theory and Computation
Shuang Zhou, Debdeep Pati, Tianying Wang, Yun Yang, Raymond J. Carroll
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 |
|---|---|---|---|
| model covariate measurement error (GPEV_a) instead of ignoring it (GP) | GPEV_a · Synthetic EIV (n=100, delta^2=1) | AMSE: 16$\cdot$81 -> 6$\cdot$31 | Improved |
| model measurement error (GPEV_a) instead of ignoring it (GP) | GPEV_a · Synthetic EIV (n=500, delta^2=0.01) | AMSE: 1$\cdot$80 -> 0$\cdot$13 | Improved |
| use GPEV_a instead of deconvolution kernel method | GPEV_a · Synthetic EIV (n=100, delta^2=1) | AMSE: 20$\cdot$59 -> 6$\cdot$31 | Improved |
| use Dirichlet process mixture prior on covariate density instead of component-wise normal prior (ablation) | GPEV_a · Synthetic EIV (n=100, delta^2=1) | AMSE: 20$\cdot$98 -> 6$\cdot$31 | Improved |
| use DP mixture prior instead of component-wise normal prior (ablation) | GPEV_a · Synthetic EIV (n=500, delta^2=1) | AMSE: 20$\cdot$38 -> 3$\cdot$35 | Improved |
| use DP mixture prior instead of normal prior when error is tiny (ablation) | GPEV_a · Synthetic EIV (n=500, delta^2=0.001) | density recovery | No clear effect |
| replace full GP (GPEV_f) with Fourier-basis approximation (GPEV_a) (ablation) | GPEV_a · Synthetic EIV (n=100, delta^2=0.01) | AMSE: 0$\cdot$55 -> 0$\cdot$58 | No clear effect |
| replace full GP with Fourier-basis approximation (ablation) | GPEV_a · Synthetic EIV (n=500) | time per MCMC iteration (s): 0.197 -> 0.025 | Improved |
| replace full GP with Fourier-basis approximation (ablation) | GPEV_a · Synthetic EIV (n=500, delta^2=1) | AMSE: 3$\cdot$94 -> 3$\cdot$35 | Mixed |
| use GPEV_a with fixed delta^2 vs unknown delta^2 with objective prior (ablation) | GPEV_a · Real data (treatment group) | MSE: 4.35 -> 4.54 | No clear effect |
| use GPEV_a instead of decon | GPEV_a · Real data (treatment group) | MSE: 8.23 -> 4.35 | Improved |
| use GPEV_a instead of decon | GPEV_a · Real data (control group) | MSE: 5.18 -> 3.87 | Improved |
Closest papers by meaning
- Empirical Gaussian Processes (ICML 2026 · similarity 0.88)
- Learning from Summarized Data: Gaussian Process Regression with Sample Quasi-Likelihood (AAAI 2025 · similarity 0.88)
- Scalable Gaussian-process regression and variable selection using Vecchia approximations (JMLR 2022 · similarity 0.87)
- Bayesian Optimization through Gaussian Cox Process Models for Spatio-temporal Data (ICLR 2024 · similarity 0.86)
- Turbocharging Gaussian Process Inference with Approximate Sketch-and-Project (NeurIPS 2025 · similarity 0.85)
- Robust Gaussian Processes via Relevance Pursuit (NeurIPS 2024 · similarity 0.85)
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.
