DisenTS: Disentangled Channel Evolving Pattern Modeling for Multivariate Time Series Forecasting

TL;DR

DisenTS introduces a novel framework for multivariate time series forecasting that models diverse evolving patterns through multiple specialized forecasters guided by a novel gating mechanism and similarity constraints.

Contribution

The paper proposes DisenTS, a framework that models disentangled channel patterns using multiple forecasters and a novel gating mechanism without supervision, improving pattern capturing in multivariate time series.

Findings

Outperforms existing methods on multiple datasets.

Effectively captures diverse channel patterns.

Enhances forecasting accuracy with disentangled modeling.

Abstract

Multivariate time series forecasting plays a crucial role in various real-world applications. Significant efforts have been made to integrate advanced network architectures and training strategies that enhance the capture of temporal dependencies, thereby improving forecasting accuracy. On the other hand, mainstream approaches typically utilize a single unified model with simplistic channel-mixing embedding or cross-channel attention operations to account for the critical intricate inter-channel dependencies. Moreover, some methods even trade capacity for robust prediction based on the channel-independent assumption. Nonetheless, as time series data may display distinct evolving patterns due to the unique characteristics of each channel (including multiple strong seasonalities and trend changes), the unified modeling methods could yield suboptimal results. To this end, we propose DisenTS, a tailored framework for modeling disentangled channel evolving patterns in general multivariate time series forecasting. The central idea of DisenTS is to model the potential diverse patterns within the multivariate time series data in a decoupled manner. Technically, the framework employs multiple distinct forecasting models, each tasked with uncovering a unique evolving pattern. To guide the learning process without supervision of pattern partition, we introduce a novel Forecaster Aware Gate (FAG) module that generates the routing signals adaptively according to both the forecasters' states and input series' characteristics. The forecasters' states are derived from the Linear Weight Approximation (LWA) strategy, which quantizes the complex deep neural networks into compact matrices. Additionally, the Similarity Constraint (SC) is further proposed to guide each model to specialize in an underlying pattern by minimizing the mutual information between the representations.