Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Black-box coreset 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 BB PSVI instead of PSVI | Logistic regression · webspam test, M=100 | accuracy: 90.3 \pm 0.3 -> 92.3 \pm 0.0 | Helped |
| Use BB PSVI instead of PSVI | Logistic regression · phishing test, M=100 | accuracy: 90.1 \pm 0.1 -> 90.4 \pm 0.0 | No clear effect |
| Use BB PSVI instead of PSVI | Logistic regression · adult test, M=100 | accuracy: 83.5 \pm 0.0 -> 83.5 \pm 0.0 | No clear effect |
| Add learnable alpha to BB PSVI (ablation) | LeNet on MNIST · MNIST test, M=100 | accuracy: 90.77 \pm 0.92 -> 91.28 \pm 0.74 | Helped |
| Add learnable weights to BB PSVI (ablation) | LeNet on MNIST · MNIST test, M=100 | accuracy: 85.47 \pm 0.21 -> 91.28 \pm 0.74 | Helped |
| Add learnable weights and alpha instead of using unscaled fixed weights (ablation) | BB PSVI for logistic regression · webspam test, M=100 | accuracy: 61.2 \pm 2.2 -> 92.3 \pm 0.0 | Helped |
| Add learnable weights and alpha instead of using unscaled fixed weights (ablation) | BB PSVI for logistic regression · phishing test, M=100 | accuracy: 87.3 \pm 11.5 -> 90.4 \pm 0.0 | Helped |
| Add learnable weights and alpha instead of using unscaled fixed weights (ablation) | BB PSVI for logistic regression · adult test, M=100 | accuracy: 75.8 \pm 1.2 -> 83.5 \pm 0.0 | Helped |
| Use uniform importance weights during training instead of learned importance weighting (ablation) | BB PSVI · webspam test, M=100 | accuracy: 92.3 \pm 0.0 -> 92.2 \pm 0.1 | No clear effect |
Closest papers by meaning
- Sparse Variational Inference: Bayesian Coresets from Scratch (NeurIPS 2019 · similarity 0.91)
- Variational Bayesian Pseudo-Coreset (ICLR 2025 · similarity 0.90)
- Generalized Variational Inference via Optimal Transport (AAAI 2024 · similarity 0.89)
- Nearly Dimension-Independent Convergence of Mean-Field Black-Box Variational Inference (NeurIPS 2025 · similarity 0.89)
- Batch and match: black-box variational inference with a score-based divergence (ICML 2024 · similarity 0.89)
- Efficient Mixture Learning in Black-Box 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.