C3RL: Rethinking the Combination of Channel-independence and Channel-mixing from Representation Learning

TL;DR

C3RL introduces a novel framework that combines channel-mixing and channel-independence strategies using contrastive learning, significantly improving multivariate time series forecasting accuracy and generalization.

Contribution

The paper proposes C3RL, a new representation learning approach that jointly models CM and CI strategies with a siamese network, enhancing performance and interpretability.

Findings

C3RL improves forecasting accuracy up to 81.4% for CI-based models.

C3RL enhances performance up to 76.3% for CM-based models.

Extensive experiments validate C3RL's effectiveness across seven models.

Abstract

Multivariate time series forecasting has drawn increasing attention due to its practical importance. Existing approaches typically adopt either channel-mixing (CM) or channel-independence (CI) strategies. CM strategy can capture inter-variable dependencies but fails to discern variable-specific temporal patterns. CI strategy improves this aspect but fails to fully exploit cross-variable dependencies like CM. Hybrid strategies based on feature fusion offer limited generalization and interpretability. To address these issues, we propose C3RL, a novel representation learning framework that jointly models both CM and CI strategies. Motivated by contrastive learning in computer vision, C3RL treats the inputs of the two strategies as transposed views and builds a siamese network architecture: one strategy serves as the backbone, while the other complements it. By jointly optimizing contrastive and prediction losses with adaptive weighting, C3RL balances representation and forecasting performance. Extensive experiments on seven models show that C3RL boosts the best-case performance rate to 81.4% for models based on CI strategy and to 76.3% for models based on CM strategy, demonstrating strong generalization and effectiveness.