Research map / Probabilistic and scientific modeling / Variational and generative sampling
Copula-like Variational Inference
Marcel Hirt, Petros Dellaportas, Alain Durmus
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 copula-like variational family instead of full-covariance Gaussian | variational inference · Bayesian logistic regression | ELBO: -2.97 -> -2.30 | Improved |
| add rotation to copula-like family (ablation) | copula-like VI · Bayesian logistic regression | ELBO: -2.30 -> -2.19 | Improved |
| use copula-like instead of full-covariance Gaussian | variational inference · Centred horseshoe | ELBO: -0.04 -> 0.04 | Improved |
| use mixture of 3 copula-like densities | copula-like VI · Centred horseshoe | ELBO: 0.04 -> 0.08 | Improved |
| copula-like base instead of independent copula, with IAF (ablation) | IAF variational family · UCI Energy | test RMSE: 1.30 -> 0.53 | Improved |
| copula-like base instead of independent copula, with rotation (ablation) | sparse rotation family · UCI Energy | test RMSE: 2.28 -> 0.55 | Improved |
| copula-like base instead of independent copula, with rotation (ablation) | sparse rotation family · UCI Protein | test RMSE: 4.51 -> 4.20 | Improved |
| copula-like base instead of independent copula (ablation) | rotation/IAF families · UCI Wine | test RMSE: 0.64 -> 0.64 | No clear effect |
| copula-like base instead of independent copula (ablation) | rotation/IAF families · UCI Kin8nm | test RMSE: 0.08 (0.00) -> 0.08 (0.00) | No clear effect |
| copula-like base instead of independent copula, with rotation (ablation) | sparse rotation family · MNIST | error rate: 2.90 \% -> 1.70 \% | Improved |
| copula-like base instead of independent copula, with IAF (ablation) | IAF family · MNIST | error rate: 2.88 \% -> 2.04 \% | Improved |
| add sparse rotation to copula-like density (ablation) | copula-like VI · MNIST | error rate: 1.78 \% -> 1.70 \% | Improved |
Closest papers by meaning
- Variational Bayesian dropout: pitfalls and fixes (ICML 2018 · similarity 0.87)
- The k-tied Normal Distribution: A Compact Parameterization of Gaussian Mean Field Posteriors in Bayesian Neural Networks (ICML 2020 · similarity 0.87)
- Variational Inference with Mixtures of Isotropic Gaussians (NeurIPS 2025 · similarity 0.87)
- Structured and Efficient Variational Deep Learning with Matrix Gaussian Posteriors (ICML 2016 · similarity 0.86)
- Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians (ICML 2024 · similarity 0.86)
- Variational Inference for Sequential Data with Future Likelihood Estimates (ICML 2020 · similarity 0.86)
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.
