Two Birds with One Stone: Series Saliency for Accurate and Interpretable Multivariate Time Series Forecasting

It is important yet challenging to perform accurate and interpretable time series forecasting. Though deep learning methods can boost forecasting accuracy, they often sacrifice interpretability. In this paper, we present a new scheme of series saliency to boost both accuracy and interpretability. By extracting series images from sliding windows of the time series, we design series saliency as a mixup strategy with a learnable mask between the series images and their perturbed versions. Series saliency is model agnostic and performs as an adaptive data augmentation method for training deep models. Moreover, by slightly changing the objective, we optimize series saliency to find a mask for interpretable forecasting in both feature and time dimensions. Experimental results on several real datasets demonstrate that series saliency is effective to produce accurate time-series forecasting results as well as generate temporal interpretations.

Mean-Square Convergence of Recursive Kernel Estimators of Non-Homogeneous Poisson Process Intensity Function and its Derivative

The structure of the estimators is similar to the recursive kernel estimators of a density function and its derivative. The estimators have been constructed using a single realization of Poisson process on a fixed time interval. Mean-square convergence has been proved in a scheme of series. Simulation studies have been carried out to illustrate the convergence.

Parameter Simulation Platform for Series Traction Mechanism of Heavy Machinery Based on ANSYS

Based on ANSYS software, second development language of ANSYS software (APDL, UIDL), the simulation scheme of series traction mechanism of heavy machinery was proposed. Based on this scheme, its simulation platform was developed, which includes parameter computing module, parameter optimization module and reliability analysis module. with this simulation platform, its parameter optimization and reliability analysis can be done. The study in the paper provides the simulation platform to predict its performance.