Research map / Probabilistic and scientific modeling / Variational and generative sampling
Graphical model inference: Sequential Monte Carlo meets deterministic approximations
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 loopy-belief-propagation twisting in SMC | SMC-Base on the Ising model · 16×16 Ising model | log normalizing constant estimate accuracy: N=64; N=1024 | Helped |
| Use EP-based twisting in SMC | SMC-Base on the LDA likelihood evaluation task · LDA synthetic toy model | mean-squared error of log-likelihood estimates: about half an order-of-magnitude | Helped |
| Use EP-based twisting in SMC | SMC-Base on the LDA likelihood evaluation task · PubMed Central abstracts | log-likelihood estimate: 50 particles | Helped |
| Use EP-based twisting in SMC | SMC-Base on the LDA likelihood evaluation task · 20 newsgroups | log-likelihood estimate: 50 particles | Helped |
| Use Laplace-based twisting in SMC | SMC-Base on the conditional autoregressive model · Germany conditional autoregressive model | log-likelihood estimate accuracy: N=64; N=1024 | Helped |
| Use twisting rather than the untwisted sequential decomposition | SMC on graphical models · Germany conditional autoregressive model | sensitivity to variable ordering | Helped |
Closest papers by meaning
- Synthesis of MCMC and Belief Propagation (NeurIPS 2016 · similarity 0.89)
- Exact Bayesian Inference on Discrete Models via Probability Generating Functions: A Probabilistic Programming Approach (NeurIPS 2023 · similarity 0.89)
- Inference Networks for Sequential Monte Carlo in Graphical Models (ICML 2016 · similarity 0.89)
- Probabilistic Circuits for Variational Inference in Discrete Graphical Models (NeurIPS 2020 · similarity 0.88)
- Online Variational Sequential Monte Carlo (ICML 2024 · similarity 0.87)
- Fearless Stochasticity in Expectation Propagation (NeurIPS 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.