Research map / Optimization / Optimizers
Using Large Ensembles of Control Variates 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 |
|---|---|---|---|
| add all seven control variates combined via regularized rule to base RP1 gradient (ablation) | RP1 base gradient · Ionosphere | ELBO after 500 iterations: −157.3 -> −72 | Helped |
| add all seven control variates combined via regularized rule to base RP1 gradient (ablation) | RP1 base gradient · Australian | ELBO after 500 iterations: −378.2 -> −251.8 | Helped |
| add all seven control variates combined via regularized rule to base RP1 gradient (ablation) | RP1 base gradient · Sonar | ELBO after 500 iterations: −442.4 -> −117.1 | Helped |
| use all 7 CVs instead of the best single CV c7 (ablation) | combined CVs · Ionosphere | ELBO after 500 iterations: −75.6 -> −72 | Helped |
| use all 7 CVs instead of the best single CV c7 (ablation) | combined CVs · Sonar | ELBO after 500 iterations: −120.2 -> −117.1 | Helped |
| use all 7 CVs instead of 4 weakest CVs (S4) (ablation) | combined CVs · Sonar | ELBO after 500 iterations: −270.2 -> −117.1 | Helped |
| add c5 to S5 to form S6 (five to six CVs) (ablation) | combined CVs · Australian | ELBO after 500 iterations: −255.1 -> −255 | No clear effect |
Closest papers by meaning
- Approximation Based Variance Reduction for Reparameterization Gradients (NeurIPS 2020 · similarity 0.92)
- VarGrad: A Low-Variance Gradient Estimator for Variational Inference (NeurIPS 2020 · similarity 0.91)
- Stochastic variance-reduced Gaussian variational inference on the Bures-Wasserstein manifold (ICLR 2025 · similarity 0.88)
- Variational Inference with Mixtures of Isotropic Gaussians (NeurIPS 2025 · similarity 0.88)
- Globally Convergent Variational Inference (NeurIPS 2024 · similarity 0.88)
- Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians (ICML 2024 · similarity 0.87)
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.