AI Paper Atlas

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

Calibrating Neural Simulation-Based Inference with Differentiable Coverage Probability

Maciej Falkiewicz, Naoya Takeishi, Imahn Shekhzadeh, Antoine Wehenkel, Arnaud Delaunoy, Gilles Louppe, Alexandros Kalousis

NeurIPS 2023 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Simulation-based inference · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
Add the conservativeness regularizer to NRENRE (CalNRE) · Six SBI benchmarksposterior coverageImproved
Add the conservativeness regularizer to NRENRE (CalNRE) · Six SBI benchmarksexpected log posterior densityMixed
Add the conservativeness regularizer to NPENPE (CalNPE) · Gravitational Wavesposterior coverageNo clear effect
Add the conservativeness regularizer to NPENPE (CalNPE) · Six SBI benchmarksexpected log posterior densityMixed
Use the calibration objective instead of the conservativeness objective (ablation)CalNRE and CalNPE · Six SBI benchmarksexpected log posterior densityHurt
Increase the regularizer sample count L (ablation)CalNRE · Weinbergposterior coverageImproved
Increase the regularizer sample count L (ablation)CalNRE · Weinbergexpected log posterior densityImproved
Increase regularization strength λ (ablation)CalNRE · WeinbergconservativenessImproved
Increase regularization strength λ (ablation)CalNRE · Weinbergexpected log posterior densityHurt

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