Research map / Label-efficient and robust learning / Recognition and long-tail
Dataset Condensation with Contrastive Signals
Saehyung Lee, Sanghyuk Chun, Sangwon Jung, Sangdoo Yun, Sungroh Yoon
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 |
|---|---|---|---|
| replace class-wise gradient matching with collective, cross-class gradient matching | Dataset condensation on fine-grained image classification · Automobile, Terrier, Fish, Lizard, Truck, and Insect | test accuracy | Improved |
| use class-wise gradient matching rather than collective cross-class matching | Dataset condensation · Fine-grained Automobile, Terrier, and Fish | test accuracy: 11.0 vs 12.2 | Hurt |
| add differentiable Siamese augmentation to DCC | DCC · SVHN, CIFAR-10, CIFAR-100 and fine-grained benchmarks | test accuracy | Mixed |
| add bi-level warm-up (ablation) | DCC · CIFAR-10, 1 image/class | test accuracy: 28.3 -> 32.9 | Improved |
| add bi-level warm-up (ablation) | DCC · CIFAR-10, 10 images/class | test accuracy: 49.2 -> 49.4 | Improved |
| add bi-level warm-up (ablation) | DCC · CIFAR-100, 1 image/class | test accuracy: 12.0 -> 13.3 | Improved |
| add bi-level warm-up (ablation) | DCC · CIFAR-100, 10 images/class | test accuracy: 28.5 -> 30.6 | Improved |
| replace no warm-up with simple inner-loop warm-up (ablation) | DCC · CIFAR-100, 1 image/class | test accuracy: 11.98 -> 13.01 | Improved |
| replace simple warm-up with proposed bi-level warm-up (ablation) | DCC · CIFAR-100, 1 image/class | test accuracy: 13.01 -> 13.27 | Improved |
| replace no warm-up with simple inner-loop warm-up (ablation) | DCC · CIFAR-100, 10 images/class | test accuracy: 28.46 -> 29.74 | Improved |
| replace simple warm-up with proposed bi-level warm-up (ablation) | DCC · CIFAR-100, 10 images/class | test accuracy: 29.74 -> 30.59 | Improved |
Closest papers by meaning
- Elucidating the Design Space of Dataset Condensation (NeurIPS 2024 · similarity 0.89)
- M3D: Dataset Condensation by Minimizing Maximum Mean Discrepancy (AAAI 2024 · similarity 0.86)
- Fetch and Forge: Efficient Dataset Condensation for Object Detection (NeurIPS 2024 · similarity 0.86)
- Diversity-Driven Synthesis: Enhancing Dataset Distillation through Directed Weight Adjustment (NeurIPS 2024 · similarity 0.86)
- Ameliorate Spurious Correlations in Dataset Condensation (ICML 2024 · similarity 0.86)
- CondTSF: One-line Plugin of Dataset Condensation for Time Series Forecasting (NeurIPS 2024 · similarity 0.85)
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.
