Structured Noise Modeling for Enhanced Time-Series Forecasting

TL;DR

This paper presents a structured noise modeling framework that enhances time-series forecasting accuracy and stability by capturing multi-scale temporal patterns through a learnable Gaussian Process and a refinement module.

Contribution

It introduces a novel forecast-blur-denoise approach with joint training, improving temporal fidelity and robustness in neural time-series models, adaptable to pretrained systems.

Findings

Consistent accuracy improvements across electricity, traffic, and solar datasets.

Enhanced stability and interpretability of fine-scale temporal predictions.

Lightweight layer effectively improves pretrained models in limited-data scenarios.

Abstract

Time-series forecasting remains difficult in real-world settings because temporal patterns operate at multiple scales, from broad contextual trends to fast, fine-grained fluctuations that drive critical decisions. Existing neural models often struggle to represent these interacting dynamics, leading to unstable predictions and reduced reliability in downstream applications. This work introduces a forecast-blur-denoise framework that improves temporal fidelity through structured noise modeling. The approach incorporates a learnable Gaussian Process module that generates smooth, correlated perturbations, encouraging the forecasting backbone to capture long-range structure while a dedicated refinement model restores high-resolution temporal detail. Training the components jointly enables natural competence division and avoids the artifacts commonly produced by isotropic corruption methods. Experiments across electricity, traffic, and solar datasets show consistent gains in multi-horizon accuracy and stability. The modular design also allows the blur-denoise layer to operate as a lightweight enhancement for pretrained models, supporting efficient adaptation in limited-data scenarios. By strengthening the reliability and interpretability of fine-scale temporal predictions, this framework contributes to more trustworthy AI systems used in forecasting-driven decision support across energy, infrastructure, and other time-critical domains.