Learning from Polar Representation: An Extreme-Adaptive Model for Long-Term Time Series Forecasting

TL;DR

This paper introduces DAN, a novel neural network model that leverages polar representation learning and specialized mechanisms to improve long-term hydrologic streamflow forecasting, especially for extreme events.

Contribution

The paper proposes DAN, an extreme-adaptive model with polar representation learning, multi-loss refinement, and Gaussian Mixture modeling for robust long-term streamflow prediction.

Findings

DAN outperforms existing hydrologic forecasting methods.

DAN effectively captures extreme events and long-range dependencies.

The model demonstrates significant improvements on real hydrologic datasets.

Abstract

In the hydrology field, time series forecasting is crucial for efficient water resource management, improving flood and drought control and increasing the safety and quality of life for the general population. However, predicting long-term streamflow is a complex task due to the presence of extreme events. It requires the capture of long-range dependencies and the modeling of rare but important extreme values. Existing approaches often struggle to tackle these dual challenges simultaneously. In this paper, we specifically delve into these issues and propose Distance-weighted Auto-regularized Neural network (DAN), a novel extreme-adaptive model for long-range forecasting of stremflow enhanced by polar representation learning. DAN utilizes a distance-weighted multi-loss mechanism and stackable blocks to dynamically refine indicator sequences from exogenous data, while also being able to handle uni-variate time-series by employing Gaussian Mixture probability modeling to improve robustness to severe events. We also introduce Kruskal-Wallis sampling and gate control vectors to handle imbalanced extreme data. On four real-life hydrologic streamflow datasets, we demonstrate that DAN significantly outperforms both state-of-the-art hydrologic time series prediction methods and general methods designed for long-term time series prediction.