Continual Learning in Real-Life Applications
Continual Learning in Real-Life Applications
复制标题
DOI：
10.1109/lra.2022.3167736
复制
发表时间：
2022-07-01
影响因子：
5.2
通讯作者：
Maltoni, Davide
中科院分区：
文献类型：
--
作者：
Graffieti, Gabriele;Borghi, Guido;Maltoni, Davide
现有的持续学习基准仅部分解决了现实生活应用程序的复杂性，限制了学习代理的现实性。在这封信中，我们提出并重点关注以现实生活场景的常见关键元素为特征的基准，包括作为输入数据的时间排序流、短时间范围内样本的强相关性、长时间范围内的高数据分布漂移以及严重的类别不平衡。此外，我们还强制执行在线训练约束，例如需要频繁更新模型，而无法存储大量过去的数据或通过模型多次传递数据集。此外，我们引入了一种基于持续学习的新颖混合方法，其架构元素和重放内存管理在所考虑的场景中被证明是有用且有效的。进行的实验验证，包括与现有方法的比较和消融研究，证实了所提出方法的有效性和适用性。
Y Existing Continual Learning benchmarks only partially address the complexity of real-life applications, limiting the realism of learning agents. In this letter, we propose and focus on benchmarks characterized by common key elements of real-life scenarios, including temporally ordered streams as input data, strong correlation of samples in short time ranges, high data distribution drift over the long time frame, and heavy class unbalancing. Moreover, we enforce online training constraints such as the need for frequent model updates without the possibility of storing a large amount of past data or passing the dataset multiple times through the model. Besides, we introduce a novel hybrid approach based on Continual Learning, whose architectural elements and replay memory management proved to be useful and effective in the considered scenarios. The experimental validation carried out, including comparisons with existing methods and an ablation study, confirms the validity and the suitability of the proposed approach.