Research map / Time series / Causal and generative time series
Sequential Monte Carlo Learning for Time Series Structure Discovery
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 SMC structure learning instead of MCMC structure learning | Gaussian process time series structure discovery · Eight Time Series Data Library datasets | runtime at comparable forecast accuracy: 10x to 100x | Helped |
| Use SMC structure learning instead of greedy structure search | Gaussian process time series structure discovery · Eight Time Series Data Library datasets | forecast error: over 2x higher | Helped |
| Use AutoGP instead of forecasting baselines | M3 monthly · M3 monthly, 1–18 step horizons | point forecast error: improvement ≥ 14% | Helped |
| Use AutoGP instead of forecasting baselines | M3 monthly · M3 monthly, 1–18 step horizons | interval forecast error: improvement ≥ 6% | Helped |
| Use AutoGP instead of the nearest-runtime baselines | M3 monthly · M3 monthly | runtime per dataset: 34.68 sec; 21.09 sec; 36.71 sec | Mixed |
| Increase the number of auxiliary parameter proposals U | Pseudo-marginal transdimensional proposals · Illustrative Linear-to-Periodic move | acceptance probability: U ≥ 5 | Helped |
| Increase U beyond the point where acceptance is already near one | Pseudo-marginal transdimensional proposals · Illustrative Linear-to-Periodic move | expected accepts per second: U ≥ 5 | No clear effect |
Closest papers by meaning
- Scalable Structure Learning of Continuous-Time Bayesian Networks from Incomplete Data (NeurIPS 2019 · similarity 0.88)
- TimeES: Probabilistic and Deterministic Time Series Forecasting via Evolutionary Spectra (NeurIPS 2026 · similarity 0.87)
- Structured Matrix Basis for Multivariate Time Series Forecasting with Interpretable Dynamics (NeurIPS 2024 · similarity 0.87)
- Multivariate Probabilistic Time Series Forecasting with Correlated Errors (NeurIPS 2024 · similarity 0.86)
- ChronoEpilogi: Scalable Time Series Selection with Multiple Solutions (NeurIPS 2024 · similarity 0.86)
- Neural structure learning with stochastic differential equations (ICLR 2024 · 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.