Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Efficient identification of informative features in simulation-based inference
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use post-hoc likelihood marginalization instead of retraining for feature subsets | NLE with MDN (FSLM) · LGM | total time: 3.89 $\pm$ 0.80 -> 1.30 $\pm$ 0.26 | Helped |
| Use post-hoc likelihood marginalization instead of retraining for feature subsets | NLE with MDN (FSLM) · LGM | KL: 0.06 $\pm$ 0.10 -> 0.07 $\pm$ 0.16 | No clear effect |
| Use post-hoc likelihood marginalization instead of retraining for feature subsets | NLE with MDN (FSLM) · HH | total time: 94.53 $\pm$ 9.24 -> 20.48 $\pm$ 2.08 | Helped |
| Use post-hoc likelihood marginalization instead of retraining for feature subsets | NLE with MDN (FSLM) · HH | KL: 29.04 $\pm$ 8.13 -> 28.27 $\pm$ 7.40 | No clear effect |
| Remove action potential threshold (APT) (ablation) | HH posterior inference · HH | posterior uncertainty | Hurt |
| Add action potential threshold (APT) (ablation) | HH posterior inference · HH | posterior uncertainty | Helped |
| Remove action potential count (APC) (ablation) | HH posterior inference · HH | posterior uncertainty | Hurt |
| Add action potential count (APC) (ablation) | HH posterior inference · HH | posterior uncertainty | Helped |
| Remove action potential amplitude (APA) (ablation) | HH posterior inference · HH | posterior uncertainty | No clear effect |
| Add action potential amplitude (APA) (ablation) | HH posterior inference · HH | posterior uncertainty | No clear effect |
| Greedily select features by choosing the one minimizing KL to the full posterior | FSLM feature selection · HH | posterior constraint | Helped |
| Select features randomly rather than greedily | HH feature selection · HH | posterior constraint | Hurt |
Closest papers by meaning
- Simultaneous identification of models and parameters of scientific simulators (ICML 2024 · similarity 0.88)
- Multilevel neural simulation-based inference (NeurIPS 2025 · similarity 0.88)
- Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models (NeurIPS 2025 · similarity 0.88)
- Compositional simulation-based inference for time series (ICLR 2025 · similarity 0.88)
- Inverse decision-making using neural amortized Bayesian actors (ICLR 2025 · similarity 0.86)
- Robust Simulation-Based Inference under Missing Data via Neural Processes (ICLR 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.