TimeCNN: Refining Cross-Variable Interaction on Time Point for Time Series Forecasting

TL;DR

TimeCNN introduces a novel convolution-based approach for multivariate time series forecasting, effectively capturing dynamic cross-variable relationships and outperforming existing models in accuracy and efficiency.

Contribution

The paper proposes a timepoint-independent convolutional model that better captures evolving variable interactions in multivariate time series forecasting.

Findings

Outperforms state-of-the-art models on 12 datasets

Reduces computational requirements by ~60%

Speeds up inference by 3-4 times

Abstract

Time series forecasting is extensively applied across diverse domains. Transformer-based models demonstrate significant potential in modeling cross-time and cross-variable interaction. However, we notice that the cross-variable correlation of multivariate time series demonstrates multifaceted (positive and negative correlations) and dynamic progression over time, which is not well captured by existing Transformer-based models. To address this issue, we propose a TimeCNN model to refine cross-variable interactions to enhance time series forecasting. Its key innovation is timepoint-independent, where each time point has an independent convolution kernel, allowing each time point to have its independent model to capture relationships among variables. This approach effectively handles both positive and negative correlations and adapts to the evolving nature of variable relationships over time. Extensive experiments conducted on 12 real-world datasets demonstrate that TimeCNN consistently outperforms state-of-the-art models. Notably, our model achieves significant reductions in computational requirements (approximately 60.46%) and parameter count (about 57.50%), while delivering inference speeds 3 to 4 times faster than the benchmark iTransformer model