Enhancing Multivariate Time Series Forecasting with Mutual Information-driven Cross-Variable and Temporal Modeling

arXiv:2403.00869v1 Announce Type: new
Abstract: Recent advancements have underscored the impact of deep learning techniques on multivariate time series forecasting (MTSF). Generally, these techniques are bifurcated into two categories: Channel-independence and Channel-mixing approaches. Although Channel-independence methods typically yield better results, Channel-mixing could theoretically offer improvements by leveraging inter-variable correlations. Nonetheless, we argue that the integration of uncorrelated information in channel-mixing methods could curtail the potential enhancement in MTSF model performance. To substantiate this claim, we introduce the Cross-variable Decorrelation Aware …

abstract arxiv cs.lg deep learning deep learning techniques forecasting impact improvements information modeling multivariate results series stat.ml temporal time series time series forecasting type