Good Seed Makes a Good Crop: Accelerating Active Learning Using Language Modeling

Good Seed Makes a Good Crop: Accelerating Active Learning Using Language Modeling
复制标题

好种子结出好庄稼：使用语言建模加速主动学习

DOI：
--
复制
发表时间：
2011
期刊：
Annual Meeting of the Association for Computational Linguistics
影响因子：
--
通讯作者：
Martha Palmer
Martha Palmer
中科院分区：
--
文献类型：
--
作者：
Dmitriy Dligach;Martha Palmer

文献摘要

被引文献

相似文献

主动学习（AL）通常使用随机选择的小种子示例进行初始化。然而，当数据中类的分布偏斜时，可能会遗漏一些类，导致学习进度缓慢。我们的贡献是双重的：（1）我们证明了一种基于无监督语言建模的技术在选择稀有类示例方面是有效的，（2）我们使用这种技术来播种AL，并证明它会导致更高的学习率。评价是在词义消歧的背景下进行的。
Active Learning (AL) is typically initialized with a small seed of examples selected randomly. However, when the distribution of classes in the data is skewed, some classes may be missed, resulting in a slow learning progress. Our contribution is twofold: (1) we show that an unsupervised language modeling based technique is effective in selecting rare class examples, and (2) we use this technique for seeding AL and demonstrate that it leads to a higher learning rate. The evaluation is conducted in the context of word sense disambiguation.