Class Incremental Learning (CIL) aims to sequentially learn new classes while avoiding catastrophic forgetting of previous knowledge. We propose to use Masked Autoencoders (MAEs) as efficient learners for CIL. MAEs were originally designed to learn useful representations through reconstructive unsupervised learning, and they can be easily integrated with a supervised loss for classification. Moreover, MAEs can reliably reconstruct original input images from randomly selected patches, which we use to store exemplars from past tasks more efficiently for CIL. We also propose a bilateral MAE framework to learn from image-level and embedding-level fusion, which produces better-quality reconstructed images and more stable representations. Our experiments confirm that our approach performs better than the state-of-the-art on CIFAR-100, ImageNet-Subset, and ImageNet-Full. The code is available at https://github.com/scok30/MAE-CIL .

翻译：增量类学习（CIL）旨在顺序学习新类别，同时避免对先前知识的灾难性遗忘。我们提出使用掩码自编码器（MAEs）作为CIL的高效学习器。MAEs最初设计用于通过重建无监督学习获取有用表征，并可轻松与用于分类的有监督损失函数集成。此外，MAEs能够从随机选取的图像块中可靠地重建原始输入图像，我们利用这一特性更高效地存储过去任务的示例以用于CIL。我们还提出了一种双边MAE框架，通过图像级和嵌入级融合进行学习，从而生成更高质量的重建图像和更稳定的表征。实验表明，我们的方法在CIFAR-100、ImageNet-Subset和ImageNet-Full数据集上均优于现有最先进方法。代码已开源在 https://github.com/scok30/MAE-CIL 。