Research map / Probabilistic and scientific modeling / Variational and generative sampling
Annealed Flow Transport 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 |
|---|---|---|---|
| Add learned normalizing flow transport (AFT) to SMC samplers, using a diagonal affine flow on VAE latent space | AFT vs SMC · VAE latent space (binarized MNIST, difficult digit) | variance of normalizing constant estimate | Helped |
| Add learned flow (AFT) to SMC on log Gaussian Cox process with diagonal affine flow | AFT vs SMC · Log Gaussian Cox process (d=1600) | accuracy of normalizing constant estimate for given number of transitions | Helped |
| Use mean-field (per-dimension) flow family AFTmf instead of SMC on 2-D target (ablation) | AFTmf vs SMC · 2-D illustrative target | variance of normalizing constant estimator for small numbers of transitions | Hurt |
| Add learned flow with inverse autoregressive dependence (AFT) to SMC on 2-D target (ablation) | AFT vs SMC · 2-D illustrative target | variance of normalizing constant estimator | Helped |
| Use learned flow on funnel distribution | AFT vs SMC · Neal's funnel (10-d) | normalizing constant estimate performance for small number of flows/transitions | No clear effect |
| Use VI proposal instead of learned flow on funnel for small flow counts | VI vs AFT · Neal's funnel (10-d) | normalizing constant estimate performance | Helped |
Closest papers by meaning
- Sequential Controlled Langevin Diffusions (ICLR 2025 · similarity 0.88)
- NETS: A Non-equilibrium Transport Sampler (ICML 2025 · similarity 0.88)
- Annealing Flow Generative Models Towards Sampling High-Dimensional and Multi-Modal Distributions (ICML 2025 · similarity 0.87)
- Provable Benefit of Annealed Langevin Monte Carlo for Non-log-concave Sampling (ICLR 2025 · similarity 0.87)
- Transport meets Variational Inference: Controlled Monte Carlo Diffusions (ICLR 2024 · similarity 0.86)
- Sampling in Unit Time with Kernel Fisher-Rao Flow (ICML 2024 · 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.