Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
HNPE: Leveraging Global Parameters for Neural Posterior Estimation
Pedro L. C. Rodrigues, Thomas Moreau, Gilles Louppe, Alexandre Gramfort
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 auxiliary observations sharing the global parameter | HNPE on the product model · Motivating example | Wasserstein distance of the global-parameter posterior to ground truth | Improved |
| Add auxiliary observations sharing the gain | HNPE on the neural mass model · Simulated neural mass model | Posterior concentration around true parameters | Improved |
| Use HNPE instead of the naive stacked-context architecture | Neural mass model · Simulated neural mass model, N > 0 | Normalized Wasserstein distance to ground truth | Improved |
| Use HNPE instead of the aggregation architecture | Neural mass model · Simulated neural mass model, N > 0 | Normalized Wasserstein distance to ground truth | Improved |
| Use the aggregation architecture instead of HNPE (ablation) | Neural mass model · Simulated neural mass model | Normalized Wasserstein distance to ground truth | No clear effect |
| Replace fixed log-PSD summaries with learned YuleNet summaries (ablation) | HNPE · Neural mass model | Posterior estimation performance | No clear effect |
| Increase the number of SNPE-C refinement rounds (ablation) | HNPE · Motivating example | Sinkhorn divergence to the analytic posterior | Improved |
| Add auxiliary observations | LFVI · Motivating example | Local-parameter posterior estimation | Hurt |
| Use hierarchical ABC instead of HNPE | Motivating example · Motivating example | Posterior concentration | Hurt |
Closest papers by meaning
- Efficient identification of informative features in simulation-based inference (NeurIPS 2022 · similarity 0.88)
- Truncated proposals for scalable and hassle-free simulation-based inference (NeurIPS 2022 · similarity 0.87)
- Robust Neural Posterior Estimation and Statistical Model Criticism (NeurIPS 2022 · similarity 0.87)
- Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models (ICML 2024 · similarity 0.85)
- Probabilistic Decomposed Linear Dynamical Systems for Robust Discovery of Latent Neural Dynamics (NeurIPS 2024 · similarity 0.84)
- CoLT: The conditional localization test for assessing the accuracy of neural posterior estimates (NeurIPS 2025 · 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.
