Channel-Aware Low-Rank Adaptation in Time Series Forecasting

TL;DR

This paper introduces a channel-aware low-rank adaptation method that enhances time series forecasting models by balancing channel independence and dependence, improving robustness and expressivity across various architectures.

Contribution

It proposes a novel plug-in low-rank adaptation technique that balances channel strategies, improving model performance and flexibility in time series forecasting.

Findings

Significant performance improvements on multiple benchmarks

Enhanced robustness to distribution shifts

Flexible integration with existing models

Abstract

The balance between model capacity and generalization has been a key focus of recent discussions in long-term time series forecasting. Two representative channel strategies are closely associated with model expressivity and robustness, including channel independence (CI) and channel dependence (CD). The former adopts individual channel treatment and has been shown to be more robust to distribution shifts, but lacks sufficient capacity to model meaningful channel interactions. The latter is more expressive for representing complex cross-channel dependencies, but is prone to overfitting. To balance the two strategies, we present a channel-aware low-rank adaptation method to condition CD models on identity-aware individual components. As a plug-in solution, it is adaptable for a wide range of backbone architectures. Extensive experiments show that it can consistently and significantly improve the performance of both CI and CD models with demonstrated efficiency and flexibility. The code is available at https://github.com/tongnie/C-LoRA.