OLC-WA: Drift Aware Tuning-Free Online Classification with Weighted Average

2025-12-14

AI Summary

Real-world data streams frequently exhibit concept drift, causing significant performance degradation in online classification models; existing approaches rely on manual hyperparameter tuning and thus struggle to adapt autonomously to evolving data distributions. This paper proposes a hyperparameter-free, adaptive concept drift-aware framework that dynamically fuses old and new models via exponential weighted moving averages, with fusion weights automatically adjusted in real time based on quantified drift intensity—requiring no human intervention. Under stationary conditions, the method achieves accuracy comparable to batch-trained models (within 1–3% error); under concept drift, it outperforms state-of-the-art online methods by 10–25% in classification accuracy. The core contribution is the first fully hyperparameter-free online drift adaptation mechanism, uniquely balancing model stability and adaptivity without sacrificing precision or requiring manual configuration.

Technology Category

• Machine Learning: Auto ML and Hyperparameter Tuning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Data Mining & Knowledge Management: Data Stream Mining

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Real-world data sets often exhibit temporal dynamics characterized by evolving data distributions. Disregarding this phenomenon, commonly referred to as concept drift, can significantly diminish a model's predictive accuracy. Furthermore, the presence of hyperparameters in online models exacerbates this issue. These parameters are typically fixed and cannot be dynamically adjusted by the user in response to the evolving data distribution. This paper introduces Online Classification with Weighted Average (OLC-WA), an adaptive, hyperparameter-free online classification model equipped with an automated optimization mechanism. OLC-WA operates by blending incoming data streams with an existing base model. This blending is facilitated by an exponentially weighted moving average. Furthermore, an integrated optimization mechanism dynamically detects concept drift, quantifies its magnitude, and adjusts the model based on the observed data stream characteristics. This approach empowers the model to effectively adapt to evolving data distributions within streaming environments. Rigorous empirical evaluation across diverse benchmark datasets shows that OLC-WA achieves performance comparable to batch models in stationary environments, maintaining accuracy within 1-3%, and surpasses leading online baselines by 10-25% under drift, demonstrating its effectiveness in adapting to dynamic data streams.

Problem

• Addresses concept drift in online classification without hyperparameters
• Adapts to evolving data distributions using weighted average blending
• Automatically detects drift and adjusts model for dynamic streams

Innovation

• Weighted average blending for online classification
• Automatic concept drift detection and quantification
• Hyperparameter-free adaptive model with dynamic adjustment