Research map / Label-efficient and robust learning / Recognition and long-tail
Deep Diffusion-Invariant Wasserstein Distributional Classification
Sung Woo Park, Dong Wook Shu, Junseok Kwon
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 DeepWDC instead of ResNet | CIFAR10 classifier · CIFAR10 {θ, sc, sh, ϵ} | accuracy: 73.6 -> 88.6 | Improved |
| use DeepWDC instead of ResNet | CIFAR10 classifier · CIFAR10 {θ2, sc, sh, ϵ} | accuracy: 73.4 -> 93.0 | Improved |
| use DeepWDC instead of DenseNet | CIFAR10 classifier · CIFAR10-C | accuracy: 92.6 -> 95.8 | Improved |
| use DeepWDC instead of DenseNet | CIFAR10 classifier · CIFAR10 {θ2, cc2} | accuracy: 74.8 -> 87.7 | Improved |
| use DeepWDC instead of DGCNN | ModelNet10 classifier · ModelNet10 {T} | accuracy: 97.1 -> 96.9 | No clear effect |
| use DeepWDC instead of DGCNN | ModelNet10 classifier · ModelNet10 {T, s, ϵ} | accuracy: 83.6 -> 94.8 | Improved |
| use DeepWDC instead of DGCNN | ModelNet10 classifier · ModelNet10 {T, s, ϵ3, θ} | accuracy: 35.4 -> 55.2 | Improved |
| use DeepWDC instead of PointNet++ | ModelNet10 classifier · ModelNet10 {T, s, ϵ3, θ} | accuracy: 25.8 -> 55.2 | Improved |
| remove the diffusion invariance term, keeping only the intrinsic distance term (ablation) | DeepWDC · CIFAR10 with perturbations | accuracy | Hurt |
| vary covariance dimension d (16, 64, 128) and rank (8, 32); use d=128 rank-32 (ablation) | DeepWDC · CIFAR10 with perturbations | accuracy | Improved |
Closest papers by meaning
- Wasserstein Distributional Normalization For Robust Distributional Certification of Noisy Labeled Data (ICML 2021 · similarity 0.85)
- Conservative Wasserstein Training for Pose Estimation (ICCV 2019 · similarity 0.85)
- Flowing Datasets with Wasserstein over Wasserstein Gradient Flows (ICML 2025 · similarity 0.83)
- Assessing the quality of denoising diffusion models in Wasserstein distance: noisy score and optimal bounds (NeurIPS 2025 · similarity 0.83)
- Wasserstein Gradient Boosting: A Framework for Distribution-Valued Supervised Learning (NeurIPS 2024 · similarity 0.83)
- Sliced Wasserstein Estimation with Control Variates (ICLR 2024 · similarity 0.83)
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.
