AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Two Phase Rapid Simulation Based Inference with Differentiable Simulators

Uros Zivanovic, Rahul Srinivasan, Roberto Trotta, Andre Scaffidi

NeurIPS 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Simulation-based inference · 0 citations (OpenAlex)

Read the paper

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces Rapid Simulation Based Inference (RSBI), a diffusion-based variational method for likelihood-free Bayesian inference using Schrödinger Bridge sampling, with likelihood from a neural ratio estimator or a differentiable simulator.

Problem

Likelihood-free Bayesian inference (simulation-based inference) struggles with posterior sample efficiency when prior-to-posterior volumes are large and multi-dimensional, and with multi-modal posteriors.

Approach

A diffusion-based variational approach that builds on Schrödinger Bridge diffusion sampling to handle multi-modal posteriors, with the likelihood supplied either by a neural ratio estimator or by a differentiable simulator (two-phase design).

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use RSBI (surrogate-guided zoom + NRE-ASBS) instead of SNPESBI on gw_ringdown · gw_ringdown default prior, 1e5 simsC2ST: 0.935 -> 0.891Improved
use RSBI instead of SNPE on wide priorSBI · gw_ringdown wide priorC2ST: 0.985 -> 0.903Improved
use RSBI instead of SNPE/SNVI on Lotka-VolterraSBI · Lotka-Volterra, 1e5 simsC2ST: 0.94 -> 0.55Improved
use RSBI at very large prior volume ratioSBI · Two Moons log10 Rp ~4.3C2ST: ~1 (baselines) -> <=0.7Improved
increase prior volume on Bernoulli GLM (to 12.2)RSBI and baselines · Bernoulli GLMC2ST: 0.66 -> 0.89 (RSBI)Hurt
increase simulation budget from 1e5 to 1e6RSBI · Two Moons default priorC2ST: 0.543 -> 0.536No clear effect
seed SNPE/SNVI with Phase-1 proposal (ablation)SNPE/SNVI · sbibm tasksC2STImproved
add Phase-2 NRE-ASBS transport on top of Phase-1 proposal (ablation)RSBI · sbibm tasksC2STImproved
use RSBI on SLCP at larger priorall methods · SLCPC2ST: baselines degrade 0.1-0.2Mixed

Experimental setup

Builds on: Schrödinger Bridge diffusion sampling

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy