Research map / Probabilistic and scientific modeling / Variational and generative sampling
CRVI: Convex Relaxation for Variational Inference
Ghazal Fazelnia, John Paisley
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 CRVI convex relaxation instead of CAVI coordinate ascent | sparse Bayesian linear regression VI · 9 UCI datasets | variational objective (relative improvement) | Improved |
| Use CRVI instead of CAVI | sparse Bayesian linear regression VI · Birth Rate & Econ | running time (s): 0.281 -> 1.115 | Hurt |
| Use CRVI instead of CAVI | sparse Bayesian linear regression VI · Iris | running time (s): 0.231 -> 1.807 | Hurt |
| Use CRVI instead of CAVI | sparse Bayesian linear regression VI · Parkinson | running time (s): 0.962 -> 7.309 | Hurt |
| Use CRVI instead of CAVI | sparse Bayesian linear regression VI · WDBC | running time (s): 1.059 -> 10.766 | Hurt |
| Use CRVI instead of CAVI | sparse Bayesian linear regression VI · Year Prediction Songs | running time (s): 18.809 -> 22.050 | Hurt |
| Increase dataset size/dimension | CRVI · 9 UCI datasets | rank of solution: 1.10 to 1.98 | No clear effect |
| Use CRVI instead of K-SVD (OMP) for sparse code recovery | nonparametric factor analysis · synthetic, N=300, D=100, K=100 | fraction of correctly recovered Z entries | Improved |
| Use CRVI instead of K-SVD (OMP) | nonparametric factor analysis · synthetic, varying sparsity | fraction of correctly recovered 1's | Improved |
Closest papers by meaning
- Robust, Accurate Stochastic Optimization for Variational Inference (NeurIPS 2020 · similarity 0.91)
- Universal Boosting Variational Inference (NeurIPS 2019 · similarity 0.90)
- Quantized Variational Inference (NeurIPS 2020 · similarity 0.89)
- Least squares variational inference (NeurIPS 2025 · similarity 0.88)
- Natural Gradient VI: Guarantees for Non-Conjugate Models (NeurIPS 2025 · similarity 0.88)
- Globally Convergent Variational Inference (NeurIPS 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.
