STUDD: A Student-Teacher Method for Unsupervised Concept Drift Detection

TL;DR

This paper introduces STUDD, an unsupervised concept drift detection method using a student-teacher paradigm, effective in real-world scenarios lacking true labels, and demonstrates competitive results across multiple data streams.

Contribution

The paper presents a novel unsupervised concept drift detection approach based on student-teacher learning, addressing limitations of loss-based methods in label-scarce environments.

Findings

Detects concept drift effectively without true labels

Shows competitive performance on 19 data streams

Outperforms some existing unsupervised methods

Abstract

Concept drift detection is a crucial task in data stream evolving environments. Most of state of the art approaches designed to tackle this problem monitor the loss of predictive models. However, this approach falls short in many real-world scenarios, where the true labels are not readily available to compute the loss. In this context, there is increasing attention to approaches that perform concept drift detection in an unsupervised manner, i.e., without access to the true labels. We propose a novel approach to unsupervised concept drift detection based on a student-teacher learning paradigm. Essentially, we create an auxiliary model (student) to mimic the behaviour of the primary model (teacher). At run-time, our approach is to use the teacher for predicting new instances and monitoring the mimicking loss of the student for concept drift detection. In a set of experiments using 19 data streams, we show that the proposed approach can detect concept drift and present a competitive behaviour relative to the state of the art approaches.