Descinet: A Hierarchical Deep Convolutional Neural Network with Skip Connection for Long Time Series Forecasting

Abstract

Time series forecasting is the process of predicting future values of a time series from knowledge of its past data. Although there are several models for making short-term predictions, the problem of long temporal sequences can still receive new contributions. Recent studies have applied Transformers-based solutions to the long-time series forecasting task and achieved good results. The problem with this technique is that the transformer architecture depends on self-attention mechanisms to effectively extract the temporal and seasonal correlations existing in the series. In this article, we used convolutional networks in a binary tree structure with skip connections between the levels of the tree that allowed greater precision and efficiency in the training. The model, was applied to 5 datasets and the experimental results show that our model achieves a significant improvement in the forecast accuracy relative to existing solutions.

Keywords: Long sequence time-series forecasting, Skips Connections, Convolutional Neural Networks (CNNs), Deep Learning