arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.16645cs.CV
收藏

面向内存受限的在线持续学习的流数据摘要

Summarizing Stream Data for Memory-Constrained Online Continual Learning

Zhejiang University(浙江大学);National University of Singapore(新加坡国立大学)

Jianyang Gu, Kai Wang, Wei Jiang, Yang You

AI总结:

本文提出SSD方法,通过蒸馏真实图像训练特征并保持训练梯度与过去任务一致性,将流数据知识摘要为高信息量样本,在受限存储下显著提升在线持续学习的重放效果。

摘要:

基于重放的方法通过从辅助记忆中复用过去的样本,证明了其在在线持续学习上的有效性。尽管在改进基于记忆的训练方案方面已做出诸多努力,但记忆中每个样本所携带的信息仍未得到充分研究。在存储空间受限的情况下,记忆的信息量对于有效重放变得至关重要。尽管一些工作设计了特定策略来选择代表性样本,但仅通过使用少量原始图像,存储空间仍未得到充分利用。为此,我们提出通过蒸馏真实图像的训练特征,将来自流数据的知识摘要到信息量更大的样本中。通过维持训练梯度的一致性以及与过去任务的关系,与原始图像相比,摘要后的样本对流数据更具代表性。在多个在线持续学习基准上进行了大量实验,以支持所提出的 SSD 方法显著增强了重放效果。我们证明,在有限的额外计算开销下,SSD 在极其受限的记忆缓冲区下为 sequential CIFAR-100 提供了超过 3% 的准确率提升。代码在 https://github.com/vimar-gu/SSD。

英文摘要:

Replay-based methods have proved their effectiveness on online continual learning by rehearsing past samples from an auxiliary memory. With many efforts made on improving training schemes based on the memory, however, the information carried by each sample in the memory remains under-investigated. Under circumstances with restricted storage space, the informativeness of the memory becomes critical for effective replay. Although some works design specific strategies to select representative samples, by only employing a small number of original images, the storage space is still not well utilized. To this end, we propose to Summarize the knowledge from the Stream Data (SSD) into more informative samples by distilling the training characteristics of real images. Through maintaining the consistency of training gradients and relationship to the past tasks, the summarized samples are more representative for the stream data compared to the original images. Extensive experiments are conducted on multiple online continual learning benchmarks to support that the proposed SSD method significantly enhances the replay effects. We demonstrate that with limited extra computational overhead, SSD provides more than 3% accuracy boost for sequential CIFAR-100 under extremely restricted memory buffer. Code in https://github.com/vimar-gu/SSD.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑