1
Return

Decompose-learn-revise: A reliability-aware framework for robust time series forecasting

delete2026-06-19
delete0
PRE
AI
X
Xinrui Zhang
Z
Zhong Shuo Chen *
DOI：10.1016/j.compeleceng.2026.111291delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• A padding-free robust decomposition mechanism is proposed to enable multi-scale temporal learning without boundary distortion. • A reliability-oriented learning pipeline integrating selective optimization and retrieval-based prediction revision is developed. • Consistent improvements across four heterogeneous domains demonstrate enhanced accuracy, robustness, and cross-domain stability.

Journal

C
Computers and Electrical Engineering
IF:
4.9
Papers:
6.7K
Citations:
1.3W

Organization

X
xi'an jiaotong-liverpool university
Scholars:
789
Papers: 432
Citations: 0
N
National University of Singapore
Scholars:
7.4W
Papers: 6.4W
Citations: 11.4W
Cited Papers

Cited Papers

Citing Papers

Citing Papers