Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
On Divergence Measures for Bayesian Pseudocoresets
Balhae Kim, Jungwon Choi, Seanie Lee, Yoonho Lee, Jung-Woo Ha, Juho Lee
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 forward KL (BPC-fKL) instead of Wasserstein (BPC-W) pseudocoreset divergence | Bayesian pseudocoreset · CIFAR10 10 ipc, HMC | accuracy: 0.3535 -> 0.4294 | Improved |
| Use forward KL instead of Wasserstein at 1 ipc | Bayesian pseudocoreset · CIFAR10 1 ipc, HMC | accuracy: 0.3435 -> 0.3354 | No clear effect |
| Use forward KL instead of reverse KL | Bayesian pseudocoreset · CIFAR10 20 ipc, HMC | accuracy: 0.3686 -> 0.4910 | Improved |
| Use pseudocoreset (BPC-fKL) instead of random coreset | Bayesian pseudocoreset · CIFAR10 10 ipc, HMC | accuracy: 0.2807 -> 0.4294 | Improved |
| Add test-time data augmentation with A-SGHMC | BPC-fKL vs BPC-W · CIFAR10 20 ipc | accuracy: 0.4910 -> 0.5153 (fKL), 0.4546 -> 0.5691 (W) | Mixed |
| Use BPC-fKL instead of BPC-W for memory and time | Bayesian pseudocoreset · CIFAR10 computational cost | GPU memory and iteration time | Improved |
| Use BPC-fKL instead of BPC-W on corrupted inputs | Bayesian pseudocoreset · CIFAR10-C | accuracy and NLL | Improved |
| Diagonal covariance Gaussian in BPC-W (ablation) | BPC-W · CIFAR10 1 ipc, A-SGHMC | accuracy: 0.2934 -> 0.2959 | No clear effect |
| Diagonal covariance Gaussian in BPC-W (ablation) | BPC-W · CIFAR10 10 ipc, A-SGHMC | accuracy: 0.4890 -> 0.4848 | No clear effect |
| Increase number of Gaussian samples S (ablation) | BPC-fKL · CIFAR10 10 ipc, HMC | accuracy | No clear effect |
| Vary Gaussian variances (too small, 0.01, larger) (ablation) | BPC-fKL · CIFAR10 10 ipc, HMC | accuracy | Mixed |
| Scale to larger pseudocoresets (50 and 100 ipc) | BPC-fKL · CIFAR10, HMC | accuracy: 0.3922 -> 0.5557 | Improved |
Closest papers by meaning
- Rethinking Dataset Distillation for Classification: Do Distilled Sets Outperform Coresets? (NeurIPS 2026 · similarity 0.88)
- Dataset Distillation for Pre-Trained Self-Supervised Vision Models (NeurIPS 2025 · similarity 0.87)
- Boost Self-Supervised Dataset Distillation via Parameterization, Predefined Augmentation, and Approximation (ICLR 2025 · similarity 0.87)
- Large Scale Dataset Distillation with Domain Shift (ICML 2024 · similarity 0.87)
- Embarrassingly Simple Dataset Distillation (ICLR 2024 · similarity 0.86)
- Provable and Efficient Dataset Distillation for Kernel Ridge Regression (NeurIPS 2024 · 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.
