About this document
Deep Learning-Based Time Series Forecasting by abdennour kebaili is a document available to read on EtoBox.
This paper reviews advancements in deep learning-based time series forecasting models from 2014 to 2024, highlighting their ability to capture correlations among time steps and variables. It discusses various algorithms, their strengths and limitations, and methods to enhance forecasting efficiency, including time series decomposition and innovative loss functions. The study also evaluates the effectiveness of these models across univariate and multivariate forecasting tasks and suggests future research dir
- Author
- abdennour kebaili
- Language
- EN