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 2011.11904cs.LG
收藏

潜在群体结构化多任务学习

Latent Group Structured Multi-task Learning

University of Tennessee, Knoxville(田纳西大学诺克斯维尔分校);University of British Columbia(不列颠哥伦比亚大学)

Xiangyu Niu, Yifan Sun, Jinyuan Sun

摘要:

在多任务学习(MTL)中,我们通过联合训练各种任务来提高关键机器学习算法的性能。当任务数量较多时,对任务结构进行建模可以进一步细化任务关系模型。例如,通常可以基于元数据或通过简单的预处理步骤(如K-means)对任务进行分组。在本文中,我们提出了我们的群体结构的潜在空间多任务学习模型,它鼓励群体结构的任务定义的先验信息。我们使用交替最小化方法来学习模型参数。在合成和真实世界的数据集上进行了实验,显示了与单任务学习(每组分别训练)和其他MTL基线相比的竞争性能。

英文摘要:

In multi-task learning (MTL), we improve the performance of key machine learning algorithms by training various tasks jointly. When the number of tasks is large, modeling task structure can further refine the task relationship model. For example, often tasks can be grouped based on metadata, or via simple preprocessing steps like K-means. In this paper, we present our group structured latent-space multi-task learning model, which encourages group structured tasks defined by prior information. We use an alternating minimization method to learn the model parameters. Experiments are conducted on both synthetic and real-world datasets, showing competitive performance over single-task learning (where each group is trained separately) and other MTL baselines.

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。

↑