Skip to yearly menu bar Skip to main content


An Efficient Dataset Condensation Plugin and Its Application to Continual Learning

Enneng Yang · Li Shen · Zhenyi Wang · Zhenyi Wang · Tongliang Liu · Guibing Guo

Great Hall & Hall B1+B2 (level 1) #213
[ ]
[ Paper [ Slides [ Poster [ OpenReview
Wed 13 Dec 8:45 a.m. PST — 10:45 a.m. PST


Dataset condensation (DC) distills a large real-world dataset into a small synthetic dataset, with the goal of training a network from scratch on the latter that performs similarly to the former. State-of-the-art (SOTA) DC methods have achieved satisfactory results through techniques such as accuracy, gradient, training trajectory, or distribution matching. However, these works all perform matching in the high-dimension pixel spaces, ignoring that natural images are usually locally connected and have lower intrinsic dimensions, resulting in low condensation efficiency. In this work, we propose a simple-yet-efficient dataset condensation plugin that matches the raw and synthetic datasets in a low-dimensional manifold. Specifically, our plugin condenses raw images into two low-rank matrices instead of parameterized image matrices. Our plugin can be easily incorporated into existing DC methods, thereby containing richer raw dataset information at limited storage costs to improve the downstream applications' performance. We verify on multiple public datasets that when the proposed plugin is combined with SOTA DC methods, the performance of the network trained on synthetic data is significantly improved compared to traditional DC methods. Moreover, when applying the DC methods as a plugin to continual learning tasks, we observed that our approach effectively mitigates catastrophic forgetting of old tasks under limited memory buffer constraints and avoids the problem of raw data privacy leakage.

Chat is not available.