Sequential Order-Robust Mamba: An Efficient Time Series Forecasting Method for Handling Channel Dependencies
SOR-Mamba is a novel time series forecasting method that leverages a regularized, unidirectional Mamba architecture and a channel correlation modeling pretraining task to effectively and efficiently capture channel dependencies in time series data, outperforming existing state-of-the-art methods.
DisenTS: A Framework for Improving Multivariate Time Series Forecasting by Modeling Disentangled Channel Evolving Patterns
DisenTS, a novel framework leveraging multiple distinct forecasting models, enhances multivariate time series forecasting by implicitly disentangling and modeling diverse channel evolving patterns.
WaveRoRA: A Novel Approach to Multivariate Time Series Forecasting Using Wavelet Transform and Rotary Route Attention
WaveRoRA is a novel model for multivariate time series forecasting that leverages wavelet transform to capture time-frequency characteristics and a novel Rotary Route Attention (RoRA) mechanism to efficiently model inter-series dependencies, achieving state-of-the-art performance with lower computational costs.
Spectral Attention for Improved Long-Range Dependency Modeling in Time Series Forecasting
This paper introduces Spectral Attention, a novel mechanism that enhances time series forecasting models by enabling them to effectively capture and utilize long-range dependencies in sequential data, leading to improved prediction accuracy.
TripCast: A Novel 2D Transformer Model for Forecasting Tourism Time Series Data using Pre-training
TripCast, a novel pre-trained 2D transformer model, effectively addresses the unique challenges of forecasting tourism time series data by considering both event time and leading time dependencies, outperforming existing methods in both in-domain and out-domain scenarios.
Frequency Interpolation for Time Series Forecasting: A Comprehensive Review and Critique of FITS
FITS, a novel time series forecasting model using frequency domain interpolation, demonstrates competitive performance with significantly reduced parameters compared to state-of-the-art models, particularly excelling in capturing periodic and seasonal patterns, but exhibiting limitations in handling trending or non-periodic behaviors.
LiNo: A Novel Time Series Forecasting Framework Using Recursive Residual Decomposition of Linear and Nonlinear Patterns
LiNo, a novel time series forecasting framework, leverages recursive residual decomposition to effectively separate and model linear and nonlinear patterns in time series data, leading to more accurate and robust predictions.
xLSTM-Mixer: An Effective Recurrent Model for Long-Term Multivariate Time Series Forecasting
xLSTM-Mixer, a novel recurrent neural network architecture, achieves state-of-the-art performance in long-term multivariate time series forecasting by effectively integrating time, variate, and multi-view mixing within an xLSTM framework.
HiPPO-KAN: A Parameter-Efficient Model for Time Series Analysis Using High-order Polynomial Projection and Kolmogorov-Arnold Networks
HiPPO-KAN, a novel neural network architecture combining High-order Polynomial Projection (HiPPO) and Kolmogorov-Arnold Networks (KAN), offers a parameter-efficient and scalable solution for time series analysis, outperforming traditional methods in accuracy and efficiency, especially for long-range forecasting.
FDF: A Novel Framework for Time Series Forecasting Using Decoupled Trend and Seasonal Modeling with Diffusion Models
The FDF framework enhances time series forecasting accuracy by decoupling trend and seasonal components, modeling each with specialized modules: PTM for trends and CDSM for seasonal patterns, outperforming existing methods on various datasets.
© 2024 by Linnk AI