Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Flow Matching Calibration for Simulation-Based Inference under Model Misspecification
Pierre-Louis Ruhlmann, Michael Arbel, Florence Forbes, Pedro L. C. Rodrigues
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 theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=0.0 | jC2ST: .613 -> .576 | Improved |
| add theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=1.0 | jC2ST: .777 -> .691 | Improved |
| add theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=2.0 | jC2ST: .864 -> .768 | Improved |
| add theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=5.0 | jC2ST: .709 -> .545 | Improved |
| add theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=0.5 | jMMD: .00051 -> .00045 | Improved |
| add theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=2.0 | jMMD: .00045 -> .00022 | Improved |
| add theta-flow T_Theta to T_X-only variant (ablation) | FMCPE · Gaussian CI toy, alpha=0.0 | jMMD: .00043 -> .00040 | Improved |
| train T_Theta and T_X jointly instead of sequentially (ablation) | FMCPE · Pendulum, Light Tunnel, Wind Tunnel | C2ST, W2, MMD | Improved |
| add T_Theta correction when learning from calibration data (vs T_X-only) (ablation) | FMCPE · Pendulum, Light Tunnel, Wind Tunnel | jC2ST, W2 | Improved |
| learn the source jointly (sequential vs T_Theta-only source) (ablation) | FMCPE · Pendulum, Light Tunnel, Wind Tunnel | jC2ST, W2 | No clear effect |
| use FMCPE instead of NPE/MF-NPE/RoPE | NPE · Pendulum / Light Tunnel / Wind Tunnel | jC2ST, W2 | Improved |
Closest papers by meaning
- Addressing Misspecification in Simulation-based Inference through Data-driven Calibration (ICML 2025 · similarity 0.91)
- Consistency Models for Scalable and Fast Simulation-Based Inference (NeurIPS 2024 · similarity 0.91)
- Inductive Domain Transfer In Misspecified Simulation-Based Inference (NeurIPS 2025 · similarity 0.89)
- Learning Likelihood-Free Reference Priors (ICML 2025 · similarity 0.87)
- MissPath-FM: Flow Matching with Structured Priors for Partially Observed Time Series (NeurIPS 2026 · similarity 0.87)
- Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models (NeurIPS 2025 · similarity 0.87)
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.
