arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1702.03614cs.MAstat.ML
收藏

具有公共潜在表示的网络上的多任务扩散自适应

Multitask diffusion adaptation over networks with common latent representations

Jie Chen, Cédric Richard, Ali H. Sayed

摘要:

以分布式和协作方式利用流数据的在线学习在广泛的应用中是有用的。近年来,这个主题受到了相当大的关注,重点是单任务和多任务场景。在单任务适应中,智能体合作跟踪共同感兴趣的目标,而在多任务适应中,智能体同时跟踪多个目标。正则化是一种有用的技术,以促进和利用在后一种情况下的任务之间的相似性。这项工作探讨了另一种方法来建模任务之间的关系,假设它们都共享一个共同的潜在特征表示。因此,提出了一种新的多任务学习公式,并以分布式在线方式为其解决方案开发了算法。我们提出了一个统一的框架来分析自适应策略的均方误差性能,并进行仿真来说明理论研究结果和潜在的应用。

英文摘要:

Online learning with streaming data in a distributed and collaborative manner can be useful in a wide range of applications. This topic has been receiving considerable attention in recent years with emphasis on both single-task and multitask scenarios. In single-task adaptation, agents cooperate to track an objective of common interest, while in multitask adaptation agents track multiple objectives simultaneously. Regularization is one useful technique to promote and exploit similarity among tasks in the latter scenario. This work examines an alternative way to model relations among tasks by assuming that they all share a common latent feature representation. As a result, a new multitask learning formulation is presented and algorithms are developed for its solution in a distributed online manner. We present a unified framework to analyze the mean-square-error performance of the adaptive strategies, and conduct simulations to illustrate the theoretical findings and potential applications.

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。

↑