Research map / Optimization / Optimizers
Robust, Accurate Stochastic Optimization for Variational 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 iterate averaging instead of last iterate (ablation) | MCSE stopping rule · Boston | k-hat: 0.55 -> 0.06 | Helped |
| use iterate averaging instead of last iterate (ablation) | MCSE stopping rule · Concrete | D_mu: 0.01 -> 0.002 | Helped |
| use iterate averaging instead of last iterate (ablation) | MCSE stopping rule · MNIST100 | k-hat: 6.94 -> 10.57 | Hurt |
| use iterate averaging instead of last iterate (ablation) | DeltaELBO stopping rule · Concrete | D_mu: 0.016 -> 0.04 | Hurt |
| use MCSE stopping rule instead of DeltaELBO | iterate averaging · Boston | k-hat: 11.2 -> 0.06 | Helped |
| use MCSE stopping rule instead of DeltaELBO | iterate averaging · Wine | k-hat: 15 -> 0.07 | Helped |
| use MCSE stopping rule instead of DeltaELBO | iterate averaging · 8-school (CP) | D_mu: 2.7 -> 0.60 | Helped |
| use MCSE stopping rule instead of DeltaELBO | iterate averaging · Radon | D_mu: 0.64 -> 0.39 | Helped |
| use MCSE stopping rule instead of DeltaELBO | iterate averaging · Radon | k-hat: 0.34 -> 0.40 | Hurt |
| use MCSE stopping rule instead of DeltaELBO | iterate averaging · MNIST100 | D_mu: 0.056 -> 0.19 | Hurt |
Closest papers by meaning
- Provable convergence guarantees for black-box variational inference (NeurIPS 2023 · similarity 0.93)
- Marginalized Stochastic Natural Gradients for Black-Box Variational Inference (ICML 2021 · similarity 0.93)
- Fast Black-box Variational Inference through Stochastic Trust-Region Optimization (NeurIPS 2017 · similarity 0.92)
- Generalized Variational Inference via Optimal Transport (AAAI 2024 · similarity 0.91)
- Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference (NeurIPS 2025 · similarity 0.90)
- Globally Convergent Variational Inference (NeurIPS 2024 · similarity 0.90)
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.