MLSP-L4.6

FILTER THEN ATTEND: IMPROVING ATTENTION-BASED TIME SERIES FORECASTING WITH SPECTRAL FILTERING

Elisha Dayag, Nhat Thanh Tran, Jack Xin, University of California Irvine, United States of America
Session:
MLSP-L4: Deep Learning Models for Time Series Forecasting Oral
Track:
Machine Learning for Signal Processing [ML]
Location:
Room 112
Presentation Time:
Tue, 5 May, 18:10 - 18:30
Presentation
Discussion
Resources
No resources available.
Session MLSP-L4
MLSP-L4.1: MOC: Mamba-based Multi-Scale One-Class Time-Series Anomaly Detection
Tiejun Wang, Rui Wang, Xudong Mou, Xu Li, Tianyu Wo, Beihang University, China; Shiru Chen, Shandong Inspur Intelligent Production Technology Co., Ltd, China; Xudong Liu, Renyu Yang, Beihang University, China
MLSP-L4.2: KDFNet: Kalman dynamic filtering network for multivariate time series forecasting
Ziqiong Li, Heyu Chai, Shengjun Liu, Central South University, China; Wenzhen Yue, Peking University, China; Xinru Liu, Central South University, China
MLSP-L4.3: KPMG: A GRAPHICAL KOOPMAN-MAMBA APPROACH FOR FINANCIAL MARKETS
Sijie Xiong, Cheng Tang, Fumiya Okubo, Kyushu University, Japan; Tsubasa Minematsu, Kyushu Institute of Technology, Japan; Yinlong Hu, Hohai University, China; Atsushi Shimada, Kyushu University, Japan
MLSP-L4.4: HIUFORMER: A HIERARCHICAL U-SHAPED TRANSFORMER WITH FREQUENCY-DIVIDED DUAL-PATH ATTENTION FOR MULTIVARIATE TIME SERIES FORECASTING
Xinyu Zhang, Zhibin Zhang, Yunong Mao, Inner Mongolia University, China
MLSP-L4.5: DyWPE: Signal-Aware Dynamic Wavelet Positional Encoding for Time Series Transformers
Habib Irani, Vangelis Metsis, Texas State University, United States of America
MLSP-L4.6: FILTER THEN ATTEND: IMPROVING ATTENTION-BASED TIME SERIES FORECASTING WITH SPECTRAL FILTERING
Elisha Dayag, Nhat Thanh Tran, Jack Xin, University of California Irvine, United States of America
Contacts