Research map / Probabilistic and scientific modeling / Variational and generative sampling
Optimally-weighted Estimators of the Maximum Mean Discrepancy for Likelihood-Free Inference
Ayush Bharti, Masha Naslidnyk, Oscar Key, Samuel Kaski, Francois-Xavier Briol
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 |
|---|---|---|---|
| Use optimally weighted rather than equally weighted IID samples | MMD estimator · g-and-k | MMD² estimation error: 2.25 -> 0.086 | Improved |
| Use optimally weighted rather than equally weighted IID samples | MMD estimator · Two moons | MMD² estimation error: 2.36 -> 0.057 | Improved |
| Use optimally weighted rather than equally weighted IID samples | MMD estimator · Bivariate Beta | MMD² estimation error: 2.13 -> 0.555 | Improved |
| Use optimally weighted rather than equally weighted IID samples | MMD estimator · MA(2) | MMD² estimation error: 2.42 -> 0.705 | Improved |
| Use optimally weighted rather than equally weighted IID samples | MMD estimator · M/G/1 queue | MMD² estimation error: 2.52 -> 1.71 | Improved |
| Use optimally weighted rather than equally weighted IID samples | MMD estimator · Lotka-Volterra | MMD² estimation error: 2.13 -> 2.04 | Improved |
| Add optimal weights to RQMC samples | MMD estimator · g-and-k | MMD² estimation error: 0.060 -> 0.059 | No clear effect |
| Add optimal weights to RQMC samples | MMD estimator · Two moons | MMD² estimation error: 0.056 -> 0.055 | No clear effect |
| Add optimal weights to RQMC samples | MMD estimator · Bivariate Beta | MMD² estimation error: 0.222 -> 0.193 | Improved |
| Add optimal weights to RQMC samples | MMD estimator · MA(2) | MMD² estimation error: 0.381 -> 0.322 | Improved |
| Add optimal weights to RQMC samples | MMD estimator · M/G/1 queue | MMD² estimation error: 0.595 -> 0.646 | Hurt |
| Add optimal weights to RQMC samples | MMD estimator · Lotka-Volterra | MMD² estimation error: 1.44 -> 1.42 | No clear effect |
Closest papers by meaning
- Minimax Estimation of Maximum Mean Discrepancy with Radial Kernels (NeurIPS 2016 · similarity 0.89)
- Two Sample Testing in High Dimension via Maximum Mean Discrepancy (JMLR 2023 · similarity 0.87)
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.87)
- Sobolev Regularized MMD Gradient Flow (NeurIPS 2026 · similarity 0.84)
- To discretize continually: Mean shift interacting particle systems for Bayesian inference (NeurIPS 2026 · similarity 0.84)
- Learning Likelihood-Free Reference Priors (ICML 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.
