DCIts -- Deep Convolutional Interpreter for time series

TL;DR

This paper presents DCIts, an interpretable deep learning model for multivariate time series forecasting that achieves high accuracy while providing transparent explanations of relevant features and optimal temporal interactions.

Contribution

The paper introduces DCIts, a novel deep learning model that enhances interpretability and performance in multivariate time series forecasting without manual supervision.

Findings

Matches or surpasses existing interpretability methods in accuracy

Effectively identifies relevant time series and lags for forecasting

Determines optimal window size and model order automatically

Abstract

We introduce an interpretable deep learning model for multivariate time series forecasting that prioritizes both predictive performance and interpretability - key requirements for understanding complex physical phenomena. Our model not only matches but often surpasses existing interpretability methods, achieving this without compromising accuracy. Through extensive experiments, we demonstrate its ability to identify the most relevant time series and lags that contribute to forecasting future values, providing intuitive and transparent explanations for its predictions. To minimize the need for manual supervision, the model is designed so one can robustly determine the optimal window size that captures all necessary interactions within the smallest possible time frame. Additionally, it effectively identifies the optimal model order, balancing complexity when incorporating higher-order terms. These advancements hold significant implications for modeling and understanding dynamic systems, making the model a valuable tool for applied and computational physicists.