TimesNet及其后续改进
TimesNet是一种基于CNN的时序预测模型，提出了周期内与周期间的理念，将时序变化基于多周期转换为二维张量 … 阅读更多
深度学习
[TSF] [GDL] A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection
基于图神经网络的时间序列分析综述详解 https://arxiv.org/abs/2307.03759 0 摘 … 阅读更多
[TSF] A Comprehensive Survey of Deep Learning for Time Series Forecasting: Architectural Diversity and Open Challenges
时序预测2025综述，pretty dope as an entry to study TSF。本文仅提供高度 … 阅读更多
[TSF] LTSF-Linear (Are Transformers Effective for Time Series Forecasting?)
本文提出了DLinear，其通道独立策略启发了PatchTST的patch-patch注意力机制。 Are T … 阅读更多
[TSF] SageFormer: Series-Aware Framework for Long-term Multivariate Time Series Forecasting
Transformer系列时序预测改进模型——SageFormer SageFormer: Series-Aw … 阅读更多