Exemplar-Free Class Incremental Learning (EFCIL) aims to learn from a sequence of tasks without having access to previous task data. In this paper, we consider the challenging Cold Start scenario in which insufficient data is available in the first task to learn a high-quality backbone. This is especially challenging for EFCIL since it requires high plasticity, which results in feature drift which is difficult to compensate for in the exemplar-free setting. To address this problem, we propose a simple and effective approach that consolidates feature representations by regularizing drift in directions highly relevant to previous tasks and employs prototypes to reduce task-recency bias. Our method, called Elastic Feature Consolidation (EFC), exploits a tractable second-order approximation of feature drift based on an Empirical Feature Matrix (EFM). The EFM induces a pseudo-metric in feature space which we use to regularize feature drift in important directions and to update Gaussian prototypes used in a novel asymmetric cross entropy loss which effectively balances prototype rehearsal with data from new tasks. Experimental results on CIFAR-100, Tiny-ImageNet, ImageNet-Subset and ImageNet-1K demonstrate that Elastic Feature Consolidation is better able to learn new tasks by maintaining model plasticity and significantly outperform the state-of-the-art.

翻译：无示例类增量学习（EFCIL）旨在无需访问先前任务数据的情况下，从一系列任务中持续学习。本文探讨了更具挑战性的冷启动场景，即首个任务中可用于学习高质量骨干网络的数据不足。这对EFCIL尤为困难，因为该方法需要较高的模型可塑性，而可塑性会导致特征漂移，在无示例设定下难以补偿。为解决此问题，我们提出一种简洁有效的方法：通过对与先前任务高度相关的方向上的特征漂移进行正则化来整合特征表示，并利用原型减轻任务近因偏差。我们提出的方法称为弹性特征整合（EFC），其基于经验特征矩阵（EFM）对特征漂移进行可处理的二阶近似。EFM在特征空间中诱导出一种伪度量，我们利用该度量对重要方向的特征漂移进行正则化，并更新高斯原型；这些原型被用于一种新颖的非对称交叉熵损失函数中，该损失函数能有效平衡原型复现与新任务数据的学习。在CIFAR-100、Tiny-ImageNet、ImageNet-Subset和ImageNet-1K数据集上的实验结果表明，弹性特征整合能通过保持模型可塑性更好地学习新任务，并显著超越现有最优方法。