MemDA: Forecasting Urban Time Series with Memory-based Drift Adaptation
This addresses the concept drift problem in urban time series forecasting for smart city applications, offering a novel method that avoids retraining and improves model adaptability, though it is incremental as it builds on existing prediction backbones.
The paper tackles the problem of concept drift in urban time series forecasting, which undermines model replicability and transferability, by proposing a memory-based drift adaptation model that encodes drift using periodicity and adjusts the model on-the-fly with a meta-dynamic network. Experiments on real-world datasets show it significantly outperforms state-of-the-art methods and generalizes well to existing prediction backbones by reducing sensitivity to distribution changes.
Urban time series data forecasting featuring significant contributions to sustainable development is widely studied as an essential task of the smart city. However, with the dramatic and rapid changes in the world environment, the assumption that data obey Independent Identically Distribution is undermined by the subsequent changes in data distribution, known as concept drift, leading to weak replicability and transferability of the model over unseen data. To address the issue, previous approaches typically retrain the model, forcing it to fit the most recent observed data. However, retraining is problematic in that it leads to model lag, consumption of resources, and model re-invalidation, causing the drift problem to be not well solved in realistic scenarios. In this study, we propose a new urban time series prediction model for the concept drift problem, which encodes the drift by considering the periodicity in the data and makes on-the-fly adjustments to the model based on the drift using a meta-dynamic network. Experiments on real-world datasets show that our design significantly outperforms state-of-the-art methods and can be well generalized to existing prediction backbones by reducing their sensitivity to distribution changes.