Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
State Space Expectation Propagation: Efficient Inference Schemes for Temporal 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 state-space EEP instead of batch EP | GP inference · Coal | NLPD: 0.922±0.11 -> 0.922±0.11 | No clear effect |
| use state-space EEP instead of batch EP | Banana · Banana | NLPD: 0.228±0.07 vs 0.216±0.10 | No clear effect |
| use EEP linearisation rather than moment matching | GP inference · Motorcycle | NLPD: 0.855±0.25 vs 0.444±0.32 | Hurt |
| use EEP linearisation rather than EP moment matching | GP inference · Audio | NLPD: -0.433±0.04 vs -0.321±0.15 | Mixed |
| add full site covariance terms through EEP | Audio GP model · Audio | NLPD: -0.433±0.04 | Helped |
| set EP power α=0 instead of α=1 (ablation) | EEP · Audio | NLPD: -0.433±0.04 -> -0.570±0.04 | Helped |
| set EP power α=0 instead of α=1 (ablation) | EEP · Rainforest | NLPD: 0.325±0.01 -> 0.309±0.01 | Helped |
| set EP power α=0 instead of α=1 (ablation) | EEP · Coal | NLPD: 0.922±0.11 -> 0.922±0.11 | No clear effect |
| set EP power α=0 instead of α=1 (ablation) | EEP · Banana | NLPD: 0.228±0.07 -> 0.229±0.07 | No clear effect |
| use EEP rather than UEP | Audio GP model · Audio | runtime: 1.661 -> 1.176 | Helped |
| use state-space inference instead of batch methods | large temporal GP models · Rainforest | applicability | Helped |
Closest papers by meaning
- Fearless Stochasticity in Expectation Propagation (NeurIPS 2024 · similarity 0.88)
- Sparse Gaussian Processes: Structured Approximations and Power-EP Revisited (NeurIPS 2025 · similarity 0.87)
- Physics-Informed Variational State-Space Gaussian Processes (NeurIPS 2024 · similarity 0.86)
- Conditioning Gaussian Processes on Almost Anything (NeurIPS 2026 · similarity 0.86)
- Exact Inference for Continuous-Time Gaussian Process Dynamics (AAAI 2024 · similarity 0.86)
- Partially Observed Trajectory Inference using Optimal Transport and a Dynamics Prior (ICLR 2025 · similarity 0.86)
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.