# The Window Dilemma: Why Concept Drift Detection is Ill-Posed

arXiv: https://arxiv.org/abs/2602.06456 · HTML: https://mlsanity.com/paper/2602.06456/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 40 references that are in the map.
- First posted on arXiv 6 Feb 2026.

## Builds on (most cited first, 10 of 10)

- [Learning under Concept Drift: A Review](https://mlsanity.com/paper/2004.05785/) (2020)
- [River: machine learning for streaming data in Python](https://mlsanity.com/paper/2012.04740/) (2020)
- [Characterizing Concept Drift](https://mlsanity.com/paper/1511.03816/) (2015)
- [From Concept Drift to Model Degradation: An Overview on Performance-Aware Drift Detectors](https://mlsanity.com/paper/2203.11070/) (2022)
- [Challenges in Benchmarking Stream Learning Algorithms with Real-world Data](https://mlsanity.com/paper/2005.00113/) (2020)
- [STUDD: A Student-Teacher Method for Unsupervised Concept Drift Detection](https://mlsanity.com/paper/2103.00903/) (2021)
- [AMF: Aggregated Mondrian Forests for Online Learning](https://mlsanity.com/paper/1906.10529/) (2019)
- [How good is the Electricity benchmark for evaluating concept drift adaptation](https://mlsanity.com/paper/1301.3524/) (2013)
- [Early Concept Drift Detection via Prediction Uncertainty](https://mlsanity.com/paper/2412.11158/) (2024)
- [Learning from Data Streams: An Overview and Update](https://mlsanity.com/paper/2212.14720/) (2022)

## Built on by (most cited first, 1 of 1)

- [Concept Drift from a Causal Perspective](https://mlsanity.com/paper/2609.25340/) (2026)

## Abstract (arXiv)

Non-stationarity of an underlying data generating process that leads to distributional changes over time is a key characteristic of Data Streams. This phenomenon, commonly referred to as Concept Drift, has been intensively studied, and Concept Drift Detectors have been established as a class of methods for detecting such changes (drifts). For the most part, Drift Detectors compare regions (windows) of the data stream and detect drift if those windows are sufficiently dissimilar. In this work, we introduce the Window Dilemma, an observation that perceived drift is a product of windowing and not necessarily the underlying data generating process. Additionally, we highlight that drift detection is ill-posed, primarily because verification of drift events are implausible in practice. We demonstrate these contributions first by an illustrative example, followed by empirical comparisons of drift detectors against a variety of alternative adaptation strategies. Our main finding is that traditional batch learning techniques often perform better than their drift-aware counterparts further bringing into question the purpose of detectors in Stream Classification.