Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Robust Hypothesis Test for Nonlinear Effect with Gaussian Processes
Jeremiah Liu, Brent Coull
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 |
|---|---|---|---|
| Estimate the RBF kernel hyperparameter by maximum likelihood | Gaussian-process interaction test · Nonlinear interaction simulation | Type I error: 0.05 -> 0.238 | Hurt |
| Use a linear or quadratic polynomial kernel for the null model | Gaussian-process interaction test · Nonlinear interaction simulation | Type I error | Hurt |
| Use a linear or quadratic polynomial kernel for the null model | Gaussian-process interaction test · Nonlinear interaction simulation | Power | Hurt |
| Use a Matérn 1/2 or Matérn 3/2 kernel when the data-generating function is smoother | Gaussian-process interaction test · Nonlinear interaction simulation with smoother true functions | Power | Hurt |
| Select the RBF kernel hyperparameter by the sample median rather than maximum likelihood | Gaussian-process interaction test · Nonlinear interaction simulation | Type I error | Improved |
| Use a neural-network kernel | Gaussian-process interaction test · Nonlinear interaction simulation | Type I error | Improved |
| Use a cross-validated ensemble of RBF kernels | Gaussian-process interaction test · Nonlinear interaction simulation across true kernels | Power | Improved |
| Use a cross-validated ensemble of neural-network kernels | Gaussian-process interaction test · Nonlinear interaction simulation across true kernels | Power | Improved |
| Use CVEK-RBF instead of CVEK-NN | Gaussian-process interaction test · Nonlinear interaction simulation | Power | Improved |
Closest papers by meaning
- A General Framework for the Analysis of Kernel-based Tests (JMLR 2024 · similarity 0.87)
- On the Robustness of Kernel Goodness-of-Fit Tests (JMLR 2025 · similarity 0.86)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.86)
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.84)
- Kernel Selection is Model Selection: A Unified Complexity-Penalised Approach for MMD Two-Sample Tests (NeurIPS 2026 · similarity 0.84)
- Permutation-Free High-Order Interaction Tests (ICML 2025 · similarity 0.84)
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.
