Research map / Probabilistic and scientific modeling / Molecules and proteins
Change Surfaces for Expressive Multidimensional Changepoints and Counterfactual Prediction
William Herlands, Daniel B. Neill, Hannes Nickisch, Andrew Gordon Wilson
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 GPCS instead of SSGP, SSGP fixed or spectral mixture kernel | GP regression · Synthetic 2D change-surface data | NMSE: 0.00078 vs 0.00200 (spectral mixture), 0.01530 (SSGP), 0.02820 (SSGP fixed) | Improved |
| add the proposed initialization procedure instead of random initialization (ablation) | GPCS · Synthetic 2D data, 30 seeds | change-surface MSE: MSE is 58% greater with random initialization; 17% larger std of sigma(w(x_i)) | Improved |
| fit a two-regime GPCS to data from a single GP (no true change) | GPCS · Synthetic single-GP data, 20 runs | std of change surface: std[sigma_1(w(x))] < 0.03 for all but two runs; one run 0.07 | No clear effect |
Closest papers by meaning
- Inferring Change Points in High-Dimensional Regression via Approximate Message Passing (JMLR 2025 · similarity 0.85)
- Spatio-temporal Bayesian On-line Changepoint Detection with Model Selection (ICML 2018 · similarity 0.84)
- An Efficient Two Step Algorithm for High Dimensional Change Point Regression Models Without Grid Search (JMLR 2019 · similarity 0.84)
- Statistically and Computationally Efficient Change Point Localization in Regression Settings (JMLR 2021 · similarity 0.84)
- Divide and Conquer Dynamic Programming: An Almost Linear Time Change Point Detection Methodology in High Dimensions (ICML 2023 · similarity 0.84)
- Uncertainty Quantification for Data-Driven Change-Point Learning via Cross-Validation (AAAI 2024 · similarity 0.83)
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.
