Abstract:The non-stationary nature of real-world Multivariate Time Series (MTS) data presents forecasting models with a formidable challenge of the time-variant distribution of time series, referred to as distribution shift. Existing studies on the distribution shift mostly adhere to adaptive normalization techniques for alleviating temporal mean and covariance shifts or time-variant modeling for capturing temporal shifts. Despite improving model generalization, these normalization-based methods often assume a time-invariant transition between outputs and inputs but disregard specific intra-/inter-series correlations, while time-variant models overlook the intrinsic causes of the distribution shift. This limits model expressiveness and interpretability of tackling the distribution shift for MTS forecasting. To mitigate such a dilemma, we present a unified Probabilistic Graphical Model to Jointly capturing intra-/inter-series correlations and modeling the time-variant transitional distribution, and instantiate a neural framework called JointPGM for non-stationary MTS forecasting. Specifically, JointPGM first employs multiple Fourier basis functions to learn dynamic time factors and designs two distinct learners: intra-series and inter-series learners. The intra-series learner effectively captures temporal dynamics by utilizing temporal gates, while the inter-series learner explicitly models spatial dynamics through multi-hop propagation, incorporating Gumbel-softmax sampling. These two types of series dynamics are subsequently fused into a latent variable, which is inversely employed to infer time factors, generate final prediction, and perform reconstruction. We validate the effectiveness and efficiency of JointPGM through extensive experiments on six highly non-stationary MTS datasets, achieving state-of-the-art forecasting performance of MTS forecasting.

Probabilistic Learning of Multivariate Time Series with Temporal Irregularity

Synergetic Learning of Heterogeneous Temporal Sequences for Multi-Horizon Probabilistic Forecasting

Temporal Multimodal Multivariate Learning

Fuzzy State-Driven Cross-Time Spatial Dependence Learning for Multivariate Time-Series Anomaly Detection

Learning Interpretable Deep State Space Model for Probabilistic Time Series Forecasting

Uncovering Multivariate Structural Dependency for Analyzing Irregularly Sampled Time Series

Multiview Spatial-Temporal Meta-Learning for Multivariate Time Series Forecasting

Robust Multivariate Time Series Forecasting against Intra- and Inter-Series Transitional Shift

A Sliding Window-Based Multi-Stage Clustering and Probabilistic Forecasting Approach for Large Multivariate Time Series Data

Order-Preserving Metric Learning for Mining Multivariate Time Series

High-Order Temporal Correlation Model Learning for Time-Series Prediction.

Multivariate Probabilistic Time Series Forecasting with Correlated Errors

Deep probabilistic graphical modeling for robust multivariate time series anomaly detection with missing data

TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series

Probabilistic Forecasting of Irregular Time Series via Conditional Flows

Bayesian Temporal Factorization for Multidimensional Time Series Prediction

Forecasting in multivariate irregularly sampled time series with missing values

Distributional Drift Adaptation with Temporal Conditional Variational Autoencoder for Multivariate Time Series Forecasting

Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing Flows

Probabilistic Imputation for Time-series Classification with Missing Data

Learning Conditional Generative Models for Temporal Point Processes