Research map / Probabilistic and scientific modeling / Variational and generative sampling
Free-Form Variational Inference for Gaussian Process State-Space Models
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 |
|---|---|---|---|
| collapsing inducing variables (FFVDCM vs FFVDM) (ablation) | FFVD SGHMC · six system identification benchmarks | test RMSE | Mixed |
| free-form coupled posterior with SGHMC (FFVDM) vs baselines | FFVD · Furnace | test RMSE: 0.571 vs VCDT 1.166 | Helped |
| collapsing inducing variables (ablation) | FFVD · Furnace | test RMSE at 1000 iterations/convergence speed | Mixed |
| particle MCMC for latent states (FFVDP) (ablation) | FFVD · Drive | test RMSE: 0.342 vs 0.673 for FFVDM | Helped |
| particle MCMC for latent states (FFVDP) (ablation) | FFVD · Ballbeam | test RMSE: 0.199 vs 0.019 for FFVDM | Hurt |
| SGHMC free-form posterior vs VCDT, NMLL | FFVD · Furnace | NMLL: -0.43 vs 7.49 | Helped |
| SGHMC free-form posterior vs VCDT, NMLL | FFVD · Ballbeam and Actuator | NMLL: Ballbeam 0.09 vs VCDT -0.65; Actuator -0.03 vs VCDT -0.36 | Hurt |
| SGHMC free-form posterior vs VCDT, running time | FFVD · Furnace | training time per 1000 iterations (s): 238.68 vs 3374.78 | Helped |
Closest papers by meaning
- Identification of Gaussian Process State Space Models (NeurIPS 2017 · similarity 0.91)
- Inference in Deep Gaussian Processes using Stochastic Gradient Hamiltonian Monte Carlo (NeurIPS 2018 · similarity 0.91)
- Physics-Informed Variational State-Space Gaussian Processes (NeurIPS 2024 · similarity 0.90)
- Online Variational Sequential Monte Carlo (ICML 2024 · similarity 0.89)
- Amortized Structured Stochastic Variational Inference for Gaussian Process Latent Variable Models (NeurIPS 2026 · similarity 0.88)
- Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians (ICML 2024 · 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.