Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Inter-domain Deep Gaussian Processes
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 |
|---|---|---|---|
| Use inter-domain Fourier-feature inducing variables in place of local inducing points | DGP with DSVI · US flight delays (N=1,000,000) | standardized RMSE: 0.932 ± 0.004 -> 0.906 ± 0.006 | Helped |
| Use inter-domain Fourier-feature inducing variables | GP with VFF · US flight delays (N=1,000,000) | standardized RMSE: 0.925 ± 0.007 -> 0.906 ± 0.006 | Helped |
| Use inter-domain deep inference instead of shallow GP with SVI | GP with SVI · US flight delays (N=1,000,000) | standardized RMSE: 0.946 ± 0.008 -> 0.906 ± 0.006 | Helped |
| Use inter-domain Fourier-feature inducing variables in place of local inducing points | DGP with DSVI · US flight delays (N=5,929,413) | standardized RMSE: 0.930 ± 0.003 -> 0.903 ± 0.002 | Helped |
| Use inter-domain Fourier-feature inducing variables | GP with VFF · US flight delays (N=5,929,413) | standardized RMSE: 0.923 ± 0.006 -> 0.903 ± 0.002 | Helped |
| Use inter-domain deep inference instead of shallow GP with SVI | GP with SVI · US flight delays (N=5,929,413) | standardized RMSE: 0.941 ± 0.005 -> 0.903 ± 0.002 | Helped |
| Add additional layers beyond two (ablation) | Inter-domain DGP with DSVI · Real-world datasets with global structure | test log-likelihood | No clear effect |
| Use an RBF kernel instead of a Matérn-3/2 kernel (ablation) | DSVI-DGP models · Regression experiments | predictive performance | Helped |
| Apply the method to datasets without global structure | Inter-domain DGP · Datasets without global structure | predictive performance | No clear effect |
Closest papers by meaning
- Spherical Inducing Features for Orthogonally-Decoupled Gaussian Processes (ICML 2023 · similarity 0.89)
- Doubly Stochastic Variational Inference for Deep Gaussian Processes (NeurIPS 2017 · similarity 0.89)
- Approximate Inference Turns Deep Networks into Gaussian Processes (NeurIPS 2019 · similarity 0.88)
- Deep Random Features for Scalable Interpolation of Spatiotemporal Data (ICLR 2025 · similarity 0.87)
- The Price of Linear Time: Error Analysis of Structured Kernel Interpolation (ICML 2025 · similarity 0.85)
- Sparse Gaussian Processes: Structured Approximations and Power-EP Revisited (NeurIPS 2025 · 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.