Research map / Probabilistic and scientific modeling / Variational and generative sampling
Automatic Reparameterisation of Probabilistic Programs
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 |
|---|---|---|---|
| Interleave centred and non-centred HMC steps | HMC on hierarchical models · German credit | ESS/∇: 1.2 ± 0.2 or 1.3 ± 0.2 -> 3.0 ± 0.2 | Helped |
| Optimise a continuous per-variable parameterisation before HMC | HMC on hierarchical models · German credit | ESS/∇: 1.2 ± 0.2 or 1.3 ± 0.2 -> 5.6 ± 0.6 | Helped |
| Optimise a continuous per-variable parameterisation before HMC | HMC on hierarchical models · German credit | ESS/∇: 3.0 ± 0.2 -> 5.6 ± 0.6 | Helped |
| Use non-centred rather than centred parameterisation when evidence is weak | The analytically studied hierarchical Gaussian model · Posterior condition number | condition number | Helped |
| Use centred rather than non-centred parameterisation when evidence is strong | The analytically studied hierarchical Gaussian model · Posterior condition number | condition number | Helped |
Closest papers by meaning
- Sound Abstraction and Decomposition of Probabilistic Programs (ICML 2018 · similarity 0.89)
- Divide, Conquer, and Combine: a New Inference Strategy for Probabilistic Programs with Stochastic Support (ICML 2020 · similarity 0.89)
- Rethinking Variational Inference for Probabilistic Programs with Stochastic Support (NeurIPS 2022 · similarity 0.88)
- RefineStat: Efficient Exploration for Probabilistic Program Synthesis (ICLR 2026 · similarity 0.88)
- Rao-Blackwellised Reparameterisation Gradients (NeurIPS 2025 · similarity 0.84)
- The Silent Hyperparameter: Quantifying the Impact of Inference Backends on LLM Reproducibility (NeurIPS 2026 · similarity 0.84)
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.