Curriculum Coarse-to-Fine Selection for High-IPC Dataset Distillation
This work addresses a specific bottleneck in dataset distillation for machine learning practitioners, offering an incremental improvement over existing combination paradigms.
The paper tackles the problem of dataset distillation effectiveness decaying in high-IPC settings by introducing a curriculum coarse-to-fine selection method, which improves state-of-the-art accuracy by up to +6.6% on CIFAR-10 and achieves 60.2% test accuracy on Tiny-ImageNet with only 0.3% degradation compared to full-dataset training.
Dataset distillation (DD) excels in synthesizing a small number of images per class (IPC) but struggles to maintain its effectiveness in high-IPC settings. Recent works on dataset distillation demonstrate that combining distilled and real data can mitigate the effectiveness decay. However, our analysis of the combination paradigm reveals that the current one-shot and independent selection mechanism induces an incompatibility issue between distilled and real images. To address this issue, we introduce a novel curriculum coarse-to-fine selection (CCFS) method for efficient high-IPC dataset distillation. CCFS employs a curriculum selection framework for real data selection, where we leverage a coarse-to-fine strategy to select appropriate real data based on the current synthetic dataset in each curriculum. Extensive experiments validate CCFS, surpassing the state-of-the-art by +6.6\% on CIFAR-10, +5.8\% on CIFAR-100, and +3.4\% on Tiny-ImageNet under high-IPC settings. Notably, CCFS achieves 60.2\% test accuracy on ResNet-18 with a 20\% compression ratio of Tiny-ImageNet, closely matching full-dataset training with only 0.3\% degradation. Code: https://github.com/CYDaaa30/CCFS.