# Online Drift Detection with Maximum Concept Discrepancy

Source: [Online Drift Detection with Maximum Concept Discrepancy](<https://www.sota2.com/research/paper/68fa74001e209131cea5f254>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T03:00:03.231Z

Authors: Ke Wan, Yi Liang, Susik Yoon

Published: 2024-07-07T13:57:50.000Z

## Abstract

Continuous learning from an immense volume of data streams becomes exceptionally critical in the internet era. However, data streams often do not conform to the same distribution over time, leading to a phenomenon called concept drift. Since a fixed static model is unreliable for inferring concept-drifted data streams, establishing an adaptive mechanism for detecting concept drift is crucial. Current methods for concept drift detection primarily assume that the labels or error rates of downstream models are given and/or underlying statistical properties exist in data streams. These approaches, however, struggle to address high-dimensional data streams with intricate irregular distribution shifts, which are more prevalent in real-world scenarios. In this paper, we propose MCD-DD, a novel concept drift detection method based on maximum concept discrepancy, inspired by the maximum mean discrepancy. Our method can adaptively identify varying forms of concept drift by contrastive learning of concept embeddings without relying on labels or statistical properties. With thorough experiments under synthetic and real-world scenarios, we demonstrate that the proposed method outperforms existing baselines in identifying concept drifts and enables qualitative analysis with high explainability.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Drift Detection on Shuttle Real](<https://www.sota2.com/research/sota/drift-detection-on-shuttle-real.md>): Drift Detection; Shuttle Real; Accuracy: 50.02.

- [Drift Detection on 2D-2G-M Synthetic](<https://www.sota2.com/research/sota/drift-detection-on-2d-2g-m-synthetic.md>): Drift Detection; 2D-2G-M Synthetic; Accuracy: 50.02.

- [Drift Detection on Noaa Real](<https://www.sota2.com/research/sota/drift-detection-on-noaa-real.md>): Drift Detection; Noaa Real; Accuracy: 50.02.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2407.05375>)

[PDF](<https://arxiv.org/pdf/2407.05375.pdf>)
