Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Towards Reliable Simulation-Based Inference with Balanced Neural Ratio Estimation
Arnaud Delaunoy, Joeri Hermans, François Rozet, Antoine Wehenkel, Gilles Louppe
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 balancing-condition penalty (lambda=100) to NRE training (BNRE) | NRE · Weinberg, SIR, M/G/1, Lotka-Volterra, GW, SLCP | expected coverage | Improved |
| add balancing penalty | NRE · all benchmarks, small simulation budgets | expected log posterior density | Hurt |
| add balancing penalty | NRE · Weinberg | posterior bias | Hurt |
| add balancing penalty | NRE · Weinberg | posterior variance | Improved |
| increase lambda from small values up to about 25 (ablation) | BNRE · Weinberg, budget 1024 | coverage AUC: lambda ≈ 25.0 | Improved |
| increase lambda to about 1000 or more (ablation) | BNRE · Weinberg, budget 1024 | posterior quality | Hurt |
| increase simulation budget from 2^10 to 2^17 | NRE and BNRE · benchmarks | coverage AUC | Improved |
Closest papers by meaning
- Direct Amortized Likelihood Ratio Estimation (AAAI 2024 · similarity 0.87)
- Simulation-Based Inference with Quantile Regression (ICML 2024 · similarity 0.86)
- Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models (NeurIPS 2025 · similarity 0.86)
- Multilevel neural simulation-based inference (NeurIPS 2025 · similarity 0.86)
- Compositional simulation-based inference for time series (ICLR 2025 · similarity 0.85)
- CoLT: The conditional localization test for assessing the accuracy of neural posterior estimates (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.
