Meta-Learning
Meta-Learning
复制标题
DOI：
10.1007/978-3-030-88132-0_2
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Wenwu Zhu;Xin Wang
中科院分区：
文献类型：
--
作者：
Wenwu Zhu;Xin Wang
在过去的十年里，监督学习得到了蓬勃的发展，即具有给定标签的学习任务用于模型训练。有监督学习通常依赖于大的标签数据集，并且从头开始训练具有大量参数的巨大模型。因此，对数据和计算资源的要求相对较高。然而，在许多应用程序中，收集数据很困难或成本很高，或者计算资源有限。由于缺乏训练数据，监督学习不适合这些任务，表现出较差的性能。
Last decade has witnessed a prosperous development for supervised learning, i.e., learning tasks with given labels for model training. Supervised learning usually depends on large labeled datasets and trains a huge model with a large number of parameters from scratch. Thus, the requirement for data and computing resources is relatively high. However, there are many applications where data is difficult or expensive to collect, or computing resources are limited. Since the lack of training data, supervised learning is not suitable for these tasks and shows bad performances.