Research map / Optimization / Optimizers
Provable convergence guarantees for black-box 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 the STL estimator instead of the entropy-gradient estimator | Projected SGD for Gaussian VI · Targets well approximated by the variational Gaussian | gradient-estimator noise | Helped |
| Use the STL estimator for an exactly Gaussian target | Projected SGD for Gaussian VI · Gaussian target | squared distance to optimum | Helped |
| Apply proximal or projected SGD using quadratically bounded gradient estimators | Gaussian black-box variational inference · Smooth, concave log-target | convergence rate | Helped |
Closest papers by meaning
- Provable Gradient Variance Guarantees for Black-Box Variational Inference (NeurIPS 2019 · similarity 0.94)
- Robust, Accurate Stochastic Optimization for Variational Inference (NeurIPS 2020 · similarity 0.93)
- On the Convergence of Black-Box Variational Inference (NeurIPS 2023 · similarity 0.93)
- Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference (NeurIPS 2025 · similarity 0.92)
- Globally Convergent Variational Inference (NeurIPS 2024 · similarity 0.89)
- Understanding Stochastic Natural Gradient Variational Inference (ICML 2024 · similarity 0.88)
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.