CoRe: Coherence and Relational Alignment for Multivariate Time Series Forecasting
CoRe is proposed, a model-agnostic learning objective for direct multivariate forecasting that replaces pointwise supervision with two output-space constraints: a frequency coherence loss that aligns predicted and target spectra, and a low-rank relational graph loss that matches sampled pairwise differences in a target...
Xiao-Yu Lin, Hui-Ran Duan, Yi-Ning Liu et al.
· 0 citations