潜在谱正则化用于持续学习
Latent Spectral Regularization for Continual Learning
AI总结：
针对持续学习中的灾难性遗忘，提出基于潜在空间拉普拉斯谱的几何正则化器CaSpeR-IL，促进类别分区，可结合任意排练方法提升SOTA性能。
摘要：
虽然生物智能在生命过程中随着新知识的获取而有机地成长，但人工神经网络在面对不断变化的训练数据分布时，会灾难性地遗忘。基于排练的持续学习（CL）方法已被确立为克服这一限制的多功能且可靠的解决方案；然而，突然的输入干扰和内存限制已知会改变其预测的一致性。我们通过研究学习者潜在空间的几何特征来研究这一现象，并发现不同类别的重放数据点日益混合，干扰了分类。因此，我们提出了一种几何正则化器，它对潜在空间的拉普拉斯谱施加弱要求，促进分区行为。我们的提议，称为用于增量学习的持续谱正则化器（CaSpeR-IL），可以轻松地与任何基于排练的CL方法结合，并提高了标准基准上SOTA方法的性能。
英文摘要：
While biological intelligence grows organically as new knowledge is gathered throughout life, Artificial Neural Networks forget catastrophically whenever they face a changing training data distribution. Rehearsal-based Continual Learning (CL) approaches have been established as a versatile and reliable solution to overcome this limitation; however, sudden input disruptions and memory constraints are known to alter the consistency of their predictions. We study this phenomenon by investigating the geometric characteristics of the learner's latent space and find that replayed data points of different classes increasingly mix up, interfering with classification. Hence, we propose a geometric regularizer that enforces weak requirements on the Laplacian spectrum of the latent space, promoting a partitioning behavior. Our proposal, called Continual Spectral Regularizer for Incremental Learning (CaSpeR-IL), can be easily combined with any rehearsal-based CL approach and improves the performance of SOTA methods on standard benchmarks.
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。
