<ul><li>Real-world time series data are inherently multivariate, often exhibiting complex inter-channel dependencies.</li><li>Proposed ChannelTokenFormer is a Transformer-based forecasting model that addresses challenges like channel dependency, asynchrony, and missingness in real-world scenarios.</li><li>The model is designed to capture cross-channel interactions, accommodate channel-wise asynchronous sampling, and handle missing values effectively.</li><li>Experiments on benchmark datasets and a real-world industrial dataset show that ChannelTokenFormer outperforms existing architectures in robustness and accuracy under challenging conditions.</li></ul>

Towards Robust Real-World Multivariate Time Series Forecasting: A Unified Framework for Dependency, Asynchrony, and Missingness

Discover more