Continual Traffic Forecasting via Mixture of Experts

TL;DR

This paper introduces TFMoE, a mixture of experts model designed for continual traffic forecasting that effectively handles evolving sensor networks without catastrophic forgetting.

Contribution

The paper proposes a novel mixture of experts framework for traffic forecasting that segments traffic into groups, enabling continual learning in evolving networks.

Findings

Outperforms existing methods on PEMSD3-Stream dataset

Demonstrates robustness against catastrophic forgetting

Achieves efficient adaptation to new sensors

Abstract

The real-world traffic networks undergo expansion through the installation of new sensors, implying that the traffic patterns continually evolve over time. Incrementally training a model on the newly added sensors would make the model forget the past knowledge, i.e., catastrophic forgetting, while retraining the model on the entire network to capture these changes is highly inefficient. To address these challenges, we propose a novel Traffic Forecasting Mixture of Experts (TFMoE) for traffic forecasting under evolving networks. The main idea is to segment the traffic flow into multiple homogeneous groups, and assign an expert model responsible for a specific group. This allows each expert model to concentrate on learning and adapting to a specific set of patterns, while minimizing interference between the experts during training, thereby preventing the dilution or replacement of prior knowledge, which is a major cause of catastrophic forgetting. Through extensive experiments on a real-world long-term streaming network dataset, PEMSD3-Stream, we demonstrate the effectiveness and efficiency of TFMoE. Our results showcase superior performance and resilience in the face of catastrophic forgetting, underscoring the effectiveness of our approach in dealing with continual learning for traffic flow forecasting in long-term streaming networks.