Dynamic Multi-period Experts for Online Time Series Forecasting

2026-03-09

AI Summary

This work addresses the limitations of existing online time series forecasting methods, which often treat concept drift as a monolithic phenomenon and struggle to adapt to diverse types of change. The authors propose DynaME, a novel framework that explicitly distinguishes concept drift into two categories: recurring and emerging. To handle recurring drift, DynaME employs a dynamic multi-horizon expert mechanism that matches historical periodic patterns; for emerging drift, it leverages an uncertainty-aware detection module to switch to a stable, general-purpose expert. The framework is backbone-agnostic and demonstrates consistent and significant performance gains over state-of-the-art methods across multiple benchmark datasets, confirming its effectiveness and robustness in complex, non-stationary environments.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Online Time Series Forecasting (OTSF) requires models to continuously adapt to concept drift. However, existing methods often treat concept drift as a monolithic phenomenon. To address this limitation, we first redefine concept drift by categorizing it into two distinct types: Recurring Drift, where previously seen patterns reappear, and Emergent Drift, where entirely new patterns emerge. We then propose DynaME (Dynamic Multi-period Experts), a novel hybrid framework designed to effectively address this dual nature of drift. For Recurring Drift, DynaME employs a committee of specialized experts that are dynamically fitted to the most relevant historical periodic patterns at each time step. For Emergent Drift, the framework detects high-uncertainty scenarios and shifts reliance to a stable, general expert. Extensive experiments on several benchmark datasets and backbones demonstrate that DynaME effectively adapts to both concept drifts and significantly outperforms existing baselines.

Problem

• Online Time Series Forecasting
• Concept Drift
• Recurring Drift
• Emergent Drift

Innovation

• concept drift
• online time series forecasting
• dynamic experts
• recurring drift
• emergent drift