Learning Parameter Distributions to Detect Concept Drift in Data Streams

Abstract

Data distributions in streaming environments are usually not stationary. In order to maintain a high predictive quality at all times, online learning models need to adapt to distributional changes, which are known as concept drift. The timely and robust identification of concept drift can be difficult, as we never have access to the true distribution of streaming data. In this work, we propose a novel framework for the detection of real concept drift, called ERICS. By treating the parameters of a predictive model as random variables, we show that concept drift corresponds to a change in the distribution of optimal parameters. To this end, we adopt common measures from information theory. The proposed framework is completely model-agnostic. By choosing an appropriate base model, ERICS is also capable to detect concept drift at the input level, which is a significant advantage over existing approaches. An evaluation on several synthetic and real-world data sets suggests that the proposed framework identifies concept drift more effectively and precisely than various existing works.

arXiv commentTo be published in the proceedings of the 25th International Conference on Pattern Recognition (ICPR 2020)

References in corpus (6)

• Characterizing Concept Drift
• Exponentially Weighted Moving Average Charts for Detecting Concept Drift
• Learning under Concept Drift: an Overview
• Reservoir of Diverse Adaptive Learners and Stacking Fast Hoeffding Drift Detection Methods for Evolving Data Streams
• Leveraging Model Inherent Variable Importance for Stable Online Feature Selection
• Online Semi-Supervised Concept Drift Detection with Density Estimation

Cited by in corpus (4)

• Deep Neural Networks and Tabular Data: A Survey
• On Baselines for Local Feature Attributions
• Learning from Data Streams: An Overview and Update
• Dynamic Model Tree for Interpretable Data Stream Learning