EXACFS——一种缓解灾难性遗忘的CIL方法
EXACFS -- A CIL Method to mitigate Catastrophic Forgetting
AI总结：
本文提出EXACFS方法，通过损失梯度估计特征显著性并随任务老化，结合蒸馏损失在类增量学习中平衡稳定性与可塑性，有效缓解灾难性遗忘。
摘要：
深度神经网络（DNNs）擅长从静态数据集中学习，但在数据按顺序到达的持续学习中表现不佳。灾难性遗忘，即遗忘先前已学知识的现象，是主要挑战。本文提出了指数平均类级特征显著性（EXACFS），以在类增量学习（CIL）设置中缓解此问题。通过使用损失梯度估计每个已学类别的模型特征显著性，在增量任务中逐渐老化该显著性，并通过蒸馏损失保留显著特征，EXACFS有效地平衡了记住旧知识（稳定性）与学习新知识（可塑性）。在CIFAR-100和ImageNet-100上的大量实验表明，EXACFS在获取可塑性的同时，在保持稳定性方面表现出卓越的性能。
英文摘要：
Deep neural networks (DNNS) excel at learning from static datasets but struggle with continual learning, where data arrives sequentially. Catastrophic forgetting, the phenomenon of forgetting previously learned knowledge, is a primary challenge. This paper introduces EXponentially Averaged Class-wise Feature Significance (EXACFS) to mitigate this issue in the class incremental learning (CIL) setting. By estimating the significance of model features for each learned class using loss gradients, gradually aging the significance through the incremental tasks and preserving the significant features through a distillation loss, EXACFS effectively balances remembering old knowledge (stability) and learning new knowledge (plasticity). Extensive experiments on CIFAR-100 and ImageNet-100 demonstrate EXACFS's superior performance in preserving stability while acquiring plasticity.
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。
