No abstract

Multivariate time series often contain missing values for reasons such as failures in the data collection mechanism. These missing values can complicate the analysis of time series data, and thus, imputation techniques are typically used to deal with this issue. However, the quality of the imputation directly affects the performance of subsequent tasks, especially when the missing rate is high. In this paper, we propose a selective imputation method that identifies a subset of time points with missing values to impute in a multivariate time series dataset. This selection, which will result in shorter and simpler time series, is based on both reducing the uncertainty of the imputations and representing the original time series as good as possible. The method is applied to different datasets to analyze the quality of the imputations and the performance obtained in subsequent tasks, such as supervised classification. The results show that it is not essential to impute all missing values as the optimal subset of time points can improve both the quality of the imputations and the accuracy of the classification.

No abstract

Despite their great success in many artificial intelligence tasks, deep neural networks (DNNs) still suffer from a few limitations, such as poor generalization behavior for out-of-distribution (OOD) data and the "black-box" nature. Information theory offers fresh insights to solve these challenges. In this short paper, we briefly review the recent developments in this area, and highlight our contributions.