Research map / Probabilistic and scientific modeling / Variational and generative sampling
Learning Gaussian Conditional Distributions using Neural Ratio Estimation is Hard
Pierre Glaser, Arthur Gretton
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Theoretical and empirical study showing that neural ratio estimation (NRE) is statistically inefficient when learning Gaussian conditional distributions.
Problem
Statistical efficiency of neural ratio estimation for simulation-based posterior estimation is poorly understood; the paper asks how many simulations NRE needs to obtain accurate posteriors.
Approach
Analyzes NRE's sample complexity specifically for Gaussian conditional distributions and argues, per the title, that learning these with NRE is hard (statistically inefficient). Full details are not available beyond the abstract.
Experimental setup
Builds on: neural ratio estimation
Closest papers by meaning
- Towards Reliable Simulation-Based Inference with Balanced Neural Ratio Estimation (NeurIPS 2022 · similarity 0.88)
- Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models (ICML 2024 · similarity 0.86)
- Direct Amortized Likelihood Ratio Estimation (AAAI 2024 · similarity 0.86)
- Neural Conditional Probability for Uncertainty Quantification (NeurIPS 2024 · similarity 0.85)
- Simulation-Based Inference with Quantile Regression (ICML 2024 · similarity 0.85)
- Density Ratio Estimation with Conditional Probability Paths (ICML 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.
