论文标题
与对比视觉变压器的在线持续学习
Online Continual Learning with Contrastive Vision Transformer
论文作者
论文摘要
在线持续学习（在线CL）研究从没有任务界限的在线数据流中学习顺序任务的问题，旨在适应新数据，同时减轻对过去任务的灾难性忘记。本文提出了一个框架对比视觉变压器（CVT），该框架设计了基于变压器体系结构的焦点对比度学习策略，以实现在线CL的更好稳定性 - 塑性折衷。具体来说，我们为在线CL设计了一种新的外部注意机制，该机制隐含地捕获了以前的任务信息。此外，CVT还包含每个班级的可学习焦点，这可能会积累以前班级的知识以减轻遗忘。基于可学习的重点，我们设计了对新阶级和过去阶级之间的重新平衡学习的重点损失，并巩固了以前学习的表示。此外，CVT包含一个双分类器结构，用于解耦学习当前类并平衡所有观察到的类。广泛的实验结果表明，我们的方法可以实现最新的性能，而在线基准测试中的参数甚至更少，并有效地减轻了灾难性的遗忘。
Online continual learning (online CL) studies the problem of learning sequential tasks from an online data stream without task boundaries, aiming to adapt to new data while alleviating catastrophic forgetting on the past tasks. This paper proposes a framework Contrastive Vision Transformer (CVT), which designs a focal contrastive learning strategy based on a transformer architecture, to achieve a better stability-plasticity trade-off for online CL. Specifically, we design a new external attention mechanism for online CL that implicitly captures previous tasks' information. Besides, CVT contains learnable focuses for each class, which could accumulate the knowledge of previous classes to alleviate forgetting. Based on the learnable focuses, we design a focal contrastive loss to rebalance contrastive learning between new and past classes and consolidate previously learned representations. Moreover, CVT contains a dual-classifier structure for decoupling learning current classes and balancing all observed classes. The extensive experimental results show that our approach achieves state-of-the-art performance with even fewer parameters on online CL benchmarks and effectively alleviates the catastrophic forgetting.
