Rainbow Memory: Continual Learning with a Memory of Diverse Samples

Autor:	Bang, Jihwan, Kim, Heesu, Yoo, YoungJoon, Ha, Jung-Woo, Choi, Jonghyun
Rok vydání:	2021
Předmět:	Computer Science - Computer Vision and Pattern Recognition Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	Continual learning is a realistic learning scenario for AI models. Prevalent scenario of continual learning, however, assumes disjoint sets of classes as tasks and is less realistic rather artificial. Instead, we focus on 'blurry' task boundary; where tasks shares classes and is more realistic and practical. To address such task, we argue the importance of diversity of samples in an episodic memory. To enhance the sample diversity in the memory, we propose a novel memory management strategy based on per-sample classification uncertainty and data augmentation, named Rainbow Memory (RM). With extensive empirical validations on MNIST, CIFAR10, CIFAR100, and ImageNet datasets, we show that the proposed method significantly improves the accuracy in blurry continual learning setups, outperforming state of the arts by large margins despite its simplicity. Code and data splits will be available in https://github.com/clovaai/rainbow-memory. Comment: Accepted paper at CVPR 2021
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2103.17230 Zobrazit plný text záznamu View this record from Arxiv