arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1706.01583cs.LGstat.ML
收藏

基于多任务学习的多层次数据集文档分类

Classifying Documents within Multiple Hierarchical Datasets using Multi-Task Learning

George Mason University(乔治梅森大学)

Azad Naik, Anveshi Charuvaka, Huzefa Rangwala

摘要:

多任务学习(MTL)是一种监督学习范式,它通过联合学习多个相关任务的预测模型来获得更好的泛化性能。当每个任务只有几个训练样本时,MTL在预测精度方面大大优于传统的单任务学习(STL)。在这项工作中,我们开发了一种基于MTL的方法,用于对在双概念层次结构(即DMOZ和维基百科)中存档的文档进行分类。我们通过为两个分层数据集上的每个不同类定义一个与其余二进制分类任务来解决多类分类问题。我们不是独立地为每个不同的任务学习线性判别式,而是使用MTL方法,使用非参数,惰性,最近邻方法建立数据集上不同任务之间的关系。我们还开发和评估了迁移学习(TL)方法,并将MTL(和TL)方法与标准的单任务学习和半监督学习方法进行了比较。我们的实证结果表明,我们开发的方法,特别是当有较少的训练样本每个分类任务的改进的强度。

英文摘要:

Multi-task learning (MTL) is a supervised learning paradigm in which the prediction models for several related tasks are learned jointly to achieve better generalization performance. When there are only a few training examples per task, MTL considerably outperforms the traditional Single task learning (STL) in terms of prediction accuracy. In this work we develop an MTL based approach for classifying documents that are archived within dual concept hierarchies, namely, DMOZ and Wikipedia. We solve the multi-class classification problem by defining one-versus-rest binary classification tasks for each of the different classes across the two hierarchical datasets. Instead of learning a linear discriminant for each of the different tasks independently, we use a MTL approach with relationships between the different tasks across the datasets established using the non-parametric, lazy, nearest neighbor approach. We also develop and evaluate a transfer learning (TL) approach and compare the MTL (and TL) methods against the standard single task learning and semi-supervised learning approaches. Our empirical results demonstrate the strength of our developed methods that show an improvement especially when there are fewer number of training examples per classification task.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑