Research map / Probabilistic and scientific modeling / Variational and generative sampling
Interacting Particle Markov Chain Monte Carlo
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 P = M/2 conditional nodes instead of other values of P (ablation) | iPMCMC · LGSSM, 1000 MCMC iterations | median error in mean estimate | Helped |
| Interaction between nodes (iPMCMC) instead of independent multi-start PG (P=M) (ablation) | iPMCMC · LGSSM | MSE of latent means over MCMC iterations | Helped |
| Increase total number of nodes M while keeping P=M/2 (ablation) | iPMCMC vs mPG · LGSSM | error relative to mPG | Helped |
| iPMCMC vs mPG at early positions of the state sequence | iPMCMC · LGSSM and NLSSM | normalized ESS | Helped |
| iPMCMC vs mPG at late positions of the state sequence | iPMCMC · LGSSM and NLSSM | normalized ESS | No clear effect |
| Replace iPMCMC with multi-start PIMH | mPIMH · LGSSM and NLSSM | ESS / accuracy: 7.3 | Hurt |
| Rao-Blackwellized estimator over Gibbs updates vs simultaneous batch Rao-Blackwellization (ablation) | iPMCMC · LGSSM | estimation performance | No clear effect |
| Use all particles via Rao-Blackwellization instead of only P retained trajectories | iPMCMC · all | estimator variance | Helped |
| Serialized alternatives with equal budget (single PG longer, more particles) vs iPMCMC (ablation) | iPMCMC · LGSSM | MSE of mean | Helped |
Closest papers by meaning
- Involutive MCMC: a Unifying Framework (ICML 2020 · similarity 0.89)
- Pseudo-Extended Markov chain Monte Carlo (NeurIPS 2019 · similarity 0.88)
- Parallelizing MCMC Across the Sequence Length (NeurIPS 2025 · similarity 0.88)
- Multi-Marginal Couplings for Metropolis--Hastings (NeurIPS 2026 · similarity 0.87)
- AutoStep: Locally adaptive involutive MCMC (ICML 2025 · similarity 0.85)
- Metropolis Adjusted Microcanonical Hamiltonian Monte Carlo (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.