Continual Learning (CL) investigates how to train Deep Networks on a stream of tasks without incurring forgetting. CL settings proposed in literature assume that every incoming example is paired with ground-truth annotations. However, this clashes with many real-world applications: gathering labeled data, which is in itself tedious and expensive, becomes infeasible when data flow as a stream. This work explores Continual Semi-Supervised Learning (CSSL): here, only a small fraction of labeled input examples are shown to the learner. We assess how current CL methods (e.g.: EWC, LwF, iCaRL, ER, GDumb, DER) perform in this novel and challenging scenario, where overfitting entangles forgetting. Subsequently, we design a novel CSSL method that exploits metric learning and consistency regularization to leverage unlabeled examples while learning. We show that our proposal exhibits higher resilience to diminishing supervision and, even more surprisingly, relying only on 25% supervision suffices to outperform SOTA methods trained under full supervision.
翻译:持续学习( CL) 调查如何在不忘记的情况下在一系列任务上培训深网络。 文献中提议的 CL 设置假设每个进取的示例都配有地面真相说明。 然而,这种与许多真实世界应用的冲突: 收集标签数据本身既乏味又昂贵, 当数据流作为流流时就变得不可行。 这项工作探索连续半监测学习( CSSL ): 这里, 向学习者展示的标签输入实例只有一小部分。 我们评估的是, 目前CL 方法( 例如: EWC、 LwF、 iCARL、 ER、 GDumb、 DER) 是如何在这种新颖而具有挑战性的情景下运行的, 在这种情景下过度纠缠不忘。 随后, 我们设计了一个新的 CSSL 方法, 利用标准学习和一致性规范在学习中利用未贴标签的例子。 我们显示, 我们的提案显示在降低监督能力方面表现出更高的韧性, 更令人惊讶的是, 仅依靠25%的监督力足以超越在全面监督下所训练的 SOTA方法。