arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.07067cs.LGstat.ML
收藏

利用时间序列特权信息高效学习非线性预测模型

Efficient learning of nonlinear prediction models with time-series privileged information

Chalmers University of Technology(查尔姆斯理工大学)

Bastian Jung, Fredrik D Johansson

AI总结:

本文将使用特权信息学习(LuPI)扩展至潜在动态系统的非线性预测任务,在映射已知至线性变换时提供理论保证,并在映射未知时提出基于随机特征和表示学习的算法,证实了特权时间序列信息能提升预测效率。

摘要:

在样本量有限的领域中,高效的学习算法至关重要。使用特权信息学习(LuPI)通过允许预测模型在训练时获取辅助信息(这些信息在模型使用时不可用)来提高样本效率。在最近的工作中,已经证明对于线性高斯动态系统中的预测,能够获取中间时间序列数据的 LuPI 学习者在期望上永远不会比任何无偏的经典学习者差,并且通常更好。我们对这一分析提供了新的见解,并将其推广到潜在动态系统中的非线性预测任务,将理论保证扩展到连接潜在变量和观测的映射仅已知至线性变换的情况。此外,针对该映射未知的情况,我们提出了基于随机特征和表示学习的算法。一系列实证结果证实了理论发现,并展示了在非线性预测中使用特权时间序列信息的潜力。

英文摘要:

In domains where sample sizes are limited, efficient learning algorithms are critical. Learning using privileged information (LuPI) offers increased sample efficiency by allowing prediction models access to auxiliary information at training time which is unavailable when the models are used. In recent work, it was shown that for prediction in linear-Gaussian dynamical systems, a LuPI learner with access to intermediate time series data is never worse and often better in expectation than any unbiased classical learner. We provide new insights into this analysis and generalize it to nonlinear prediction tasks in latent dynamical systems, extending theoretical guarantees to the case where the map connecting latent variables and observations is known up to a linear transform. In addition, we propose algorithms based on random features and representation learning for the case when this map is unknown. A suite of empirical results confirm theoretical findings and show the potential of using privileged time-series information in nonlinear prediction.

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。

↑