Multi-Variate Time Series Forecasting on Variable Subsets(KDD2022)

本文探讨了在存在长期数据丢失或资源领域切换时,MTSF模型的鲁棒性问题。通过实验展示了现有方法在变子集预测(VSF)中的性能下降,提出了一种非参数包装技术,能有效恢复模型性能,即使仅15%变量可用。
摘要由CSDN通过智能技术生成

本文在多变量时间序列预测(MTSF)领域制定了一种新的推理任务,称为变子集预测(VSF)在推理过程中只有一小部分变量可用。在推理过程中,由于长期的数据丢失,变量是缺失的。传感器故障)或高→低资源域在训练/测试之间切换。据我们所知,在存在此类故障时,MTSF模型的鲁棒性还没有在文献中研究过。通过广泛的评估,本文首先表明,最先进方法的性能在VSF设置中显著下降。本文提出一种非参数的包装技术,可应用于任何现有的预测模型。通过对4个数据集和5个预测模型的系统实验,表明所提出技术能够恢复模型近95%的性能,即使只有15%的原始变量存在。

 Variables are absent during inference because of longterm data loss (eg. sensor failures) or high→low-resource domain shift between train / test. 

  • 3
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
variate A multivariate time series prediction model in Python is a model that takes multiple variables as inputs to predict the future values of a time series. This can be useful in situations where the behavior of a time series is influenced by multiple factors or variables. There are several approaches to building multivariate time series prediction models in Python, including: 1. Vector Autoregression (VAR) Model: The VAR model is a popular approach for modeling multivariate time series data. It assumes that each variable in the time series is affected by its own lagged values as well as the lagged values of the other variables. 2. Long Short-Term Memory (LSTM) Network: LSTM is a type of recurrent neural network that is commonly used for modeling time series data. It can handle multivariate time series data by taking multiple input sequences and predicting multiple output sequences. 3. Convolutional Neural Network (CNN): CNNs are commonly used for image processing tasks, but they can also be used for time series prediction. In a multivariate time series prediction context, CNNs can take multiple time series as inputs and predict multiple output time series. 4. Support Vector Regression (SVR): SVR is a machine learning technique that can be used for time series prediction. It works by finding a hyperplane that best separates the input data into different classes, and then uses this hyperplane to make predictions. Overall, the choice of model will depend on the specific characteristics of the time series data and the prediction task at hand. It is important to evaluate the performance of different models using appropriate metrics and techniques such as cross-validation.

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值