NRTSI: Non-Recurrent Time Series Imputation
NRTSI: Non-Recurrent Time Series Imputation
复制标题
DOI：
10.1109/icassp49357.2023.10095054
复制
发表时间：
2021-02
期刊：
影响因子：
--
通讯作者：
Siyuan Shan;Yang Li;Junier B. Oliva
中科院分区：
文献类型：
--
作者：
Siyuan Shan;Yang Li;Junier B. Oliva
时间序列插补是理解序列数据的一项基本任务。现有的方法要么依赖于经常性的模型，遭受严重的错误复合或未能利用时序数据的层次信息，这两个严重降低性能与稀疏观测数据。在这项工作中，我们重新制定时间序列集，并提出了一种新的非经常性的插补模型，非经常性时间序列插补（NRTSI），不施加任何经常性的结构。利用集合公式，我们设计了一个原则性的和有效的分层插补过程。此外，NRTSI可以执行多模式随机插补，直接处理不规则采样的时间序列，并处理具有部分观测维度的数据。从经验上讲，我们表明，NRTSI在多个基准测试中达到了最先进的性能。
Time series imputation is a fundamental task in understanding sequential data. Existing methods either rely on recurrent models that suffer heavily from error compounding or fail to exploit the hierarchical information of temporal data, both of which degrade performance severely with sparsely observed data. In this work, we reformulate time series as sets and propose a novel non-recurrent imputation model, Non-Recurrent Time Series Imputation (NRTSI), that does not impose any recurrent structures. Taking advantage of the set formulation, we design a principled and efficient hierarchical imputation procedure. In addition, NRTSI can perform multiple-mode stochastic imputation, directly handle irregularly-sampled time series, and handle data with partially observed dimensions. Empirically, we show that NRTSI achieves state-of-the-art performance on multiple benchmarks.