Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Modular Gaussian Processes for Transfer Learning
Pablo Moreno-Muñoz, Antonio Artés, Mauricio A Álvarez
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 ModularGP instead of GPoE | distributed GP · Synthetic regression 10k | RMSE: 2.43 -> 1.56 | Improved |
| use ModularGP instead of PoE | distributed GP · Synthetic regression 100k | RMSE: 2.19 -> 1.73 | Improved |
| use ModularGP instead of RBCM at 1m points | distributed GP · Synthetic regression 1m | NLPD: 2.56 -> 2.87 | Hurt |
| use ModularGP instead of GPoE at 100k points | distributed GP · Synthetic regression 100k | NLPD: 2.73 -> 2.89 | Hurt |
| use ModularGP instead of full-data exact inference setup of prior work | distributed GP · Synthetic regression 100k | NLPD: 2.58 -> 2.89 | Hurt |
| meta-MOGP (Q=2) instead of independent modules | ModularGP · US flight, months | NLPD: -17.69% | Mixed |
| ModularGP Q=3 instead of Q=2 | ModularGP · US flight, months | NLPD: 2.51 -> 2.38 | Improved |
| ModularGP Q=4 instead of Q=2 | ModularGP · US flight, days | NLPD: 2.49 -> 2.36 | Improved |
| ModularGP instead of MOGP baseline (Q=2) | MOGP · US flight, months | NLPD: 2.58 -> 2.51 | Improved |
| ModularGP instead of MOGP baseline (Q=4) | MOGP · US flight, months | NLPD: 2.34 -> 2.39 | Hurt |
| modular GP instead of variational GP baseline | ModularGP · Banana classification | test NLPD: 7.29 -> 7.21 | Improved |
| meta-MOGP from modules, regression task | Meta-MOGP · London household | NLPD: 4.18 -> 5.49 | Hurt |
Closest papers by meaning
- Sparse Gaussian Processes with Spherical Harmonic Features (ICML 2020 · similarity 0.89)
- Orthogonally Decoupled Variational Gaussian Processes (NeurIPS 2018 · similarity 0.89)
- Sparse Gaussian Processes: Structured Approximations and Power-EP Revisited (NeurIPS 2025 · similarity 0.88)
- Sparse Gaussian Process Hyperparameters: Optimize or Integrate? (NeurIPS 2022 · similarity 0.88)
- New Bounds for Sparse Variational Gaussian Processes (ICML 2025 · similarity 0.88)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.87)
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.
