arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1412.7489stat.MLcs.LGcs.NE
收藏

多领域多任务学习的统一视角

A Unified Perspective on Multi-Domain and Multi-Task Learning

Queen Mary, University of London(伦敦大学玛丽女王学院)

Yongxin Yang, Timothy M. Hospedales

摘要:

在本文中,我们提供了一个新的基于神经网络的多任务学习(MTL)和多域学习(MDL)的观点。通过引入语义描述符的概念,该框架统一了MDL和MTL,并通过将它们解释为构建语义描述符的不同方式来包含各种经典和最近的MTL/MDL算法。我们的解释为zero-shot learning(ZML)提供了一个替代管道,其中可以在没有训练数据的情况下构建新类的模型。此外,它导致了一个新的和实际相关的问题设置的zero-shot域自适应(ZHADA),这是类似于ZHML,但新的域:一个模型,一个看不见的域可以生成其语义描述符。在这一系列问题的实验表明,我们的框架优于各种替代品。

英文摘要:

In this paper, we provide a new neural-network based perspective on multi-task learning (MTL) and multi-domain learning (MDL). By introducing the concept of a semantic descriptor, this framework unifies MDL and MTL as well as encompassing various classic and recent MTL/MDL algorithms by interpreting them as different ways of constructing semantic descriptors. Our interpretation provides an alternative pipeline for zero-shot learning (ZSL), where a model for a novel class can be constructed without training data. Moreover, it leads to a new and practically relevant problem setting of zero-shot domain adaptation (ZSDA), which is the analogous to ZSL but for novel domains: A model for an unseen domain can be generated by its semantic descriptor. Experiments across this range of problems demonstrate that our framework outperforms a variety of alternatives.

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。

↑