# Detecting Concept Drift With Neural Network Model Uncertainty

arXiv: https://arxiv.org/abs/2107.01873 · HTML: https://mlsanity.com/paper/2107.01873/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 51 references that are in the map.
- First posted on arXiv 5 Jul 2021.

## Builds on (most cited first, 15 of 15)

- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Evidential Deep Learning to Quantify Classification Uncertainty](https://mlsanity.com/paper/1806.01768/) (2018)
- [Weight Uncertainty in Neural Networks](https://mlsanity.com/paper/1505.05424/) (2015)
- [Learning under Concept Drift: A Review](https://mlsanity.com/paper/2004.05785/) (2020)
- [Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks](https://mlsanity.com/paper/1502.05336/) (2015)
- [Scikit-Multiflow: A Multi-output Streaming Framework](https://mlsanity.com/paper/1807.04662/) (2018)
- [Challenges in Benchmarking Stream Learning Algorithms with Real-world Data](https://mlsanity.com/paper/2005.00113/) (2020)
- [Reactive Soft Prototype Computing for Concept Drift Streams](https://mlsanity.com/paper/2007.05432/) (2020)
- [DEAL: Deep Evidential Active Learning for Image Classification](https://mlsanity.com/paper/2007.11344/) (2020)
- [Inherent Brain Segmentation Quality Control from Fully ConvNet Monte Carlo Sampling](https://mlsanity.com/paper/1804.07046/) (2018)
- [Switching Scheme: A Novel Approach for Handling Incremental Concept Drift in Real-World Data Sets](https://mlsanity.com/paper/2011.02738/) (2020)

## Built on by (most cited first, 22 of 22)

- [Machine Learning Operations (MLOps): Overview, Definition, and Architecture](https://mlsanity.com/paper/2205.02302/) (2022)
- [Active learning for data streams: a survey](https://mlsanity.com/paper/2302.08893/) (2023)
- [Active Test-Time Adaptation: Theoretical Analyses and An Algorithm](https://mlsanity.com/paper/2404.05094/) (2024)
- [Confidence-based Estimators for Predictive Performance in Model Monitoring](https://mlsanity.com/paper/2407.08649/) (2024)
- [Autoregressive based Drift Detection Method](https://mlsanity.com/paper/2203.04769/) (2022)
- [QUTE: Quantifying Uncertainty in TinyML with Early-exit-assisted ensembles for model-monitoring](https://mlsanity.com/paper/2404.12599/) (2024)
- [On The Reliability Of Machine Learning Applications In Manufacturing Environments](https://mlsanity.com/paper/2112.06986/) (2021)
- [DetAIL : A Tool to Automatically Detect and Analyze Drift In Language](https://mlsanity.com/paper/2211.04250/) (2022)
- [Cluster Analysis and Concept Drift Detection in Malware](https://mlsanity.com/paper/2502.14135/) (2025)
- [Measuring the Confidence of Traffic Forecasting Models: Techniques, Experimental Comparison and Guidelines towards Their Actionability](https://mlsanity.com/paper/2210.16049/) (2022)
- [Securing Reliability: A Brief Overview on Enhancing In-Context Learning for Foundation Models](https://mlsanity.com/paper/2402.17671/) (2024)
- [Detecting Concept Drift in Neural Networks Using Chi-squared Goodness of Fit Testing](https://mlsanity.com/paper/2505.04318/) (2025)
- [On the Connection between Concept Drift and Uncertainty in Industrial Artificial Intelligence](https://mlsanity.com/paper/2303.07940/) (2023)
- [Towards Practicable Sequential Shift Detectors](https://mlsanity.com/paper/2307.14758/) (2023)
- [Reliability Quantification of Deep Reinforcement Learning-based Control](https://mlsanity.com/paper/2309.16977/) (2023)
- [An Empirical Study of Uncertainty Estimation Techniques for Detecting Drift in Data Streams](https://mlsanity.com/paper/2311.13374/) (2023)
- [Concept Drift Detection using Ensemble of Integrally Private Models](https://mlsanity.com/paper/2406.04903/) (2024)
- [Time to Retrain? Detecting Concept Drifts in Machine Learning Systems](https://mlsanity.com/paper/2410.09190/) (2024)
- [A Multi-Level Strategy for Deepfake Content Moderation under EU Regulation](https://mlsanity.com/paper/2507.08879/) (2025)
- [An Adaptive Sampling Framework for Detecting Localized Concept Drift under Label Scarcity](https://mlsanity.com/paper/2511.02452/) (2025)
- [McNdroid: A Longitudinal Multimodal Benchmark for Robust Drift Detection in Android Malware](https://mlsanity.com/paper/2605.06894/) (2026)
- [Concept Drift from a Causal Perspective](https://mlsanity.com/paper/2609.25340/) (2026)

## Abstract (arXiv)

Deployed machine learning models are confronted with the problem of changing data over time, a phenomenon also called concept drift. While existing approaches of concept drift detection already show convincing results, they require true labels as a prerequisite for successful drift detection. Especially in many real-world application scenarios-like the ones covered in this work-true labels are scarce, and their acquisition is expensive. Therefore, we introduce a new algorithm for drift detection, Uncertainty Drift Detection (UDD), which is able to detect drifts without access to true labels. Our approach is based on the uncertainty estimates provided by a deep neural network in combination with Monte Carlo Dropout. Structural changes over time are detected by applying the ADWIN technique on the uncertainty estimates, and detected drifts trigger a retraining of the prediction model. In contrast to input data-based drift detection, our approach considers the effects of the current input data on the properties of the prediction model rather than detecting change on the input data only (which can lead to unnecessary retrainings). We show that UDD outperforms other state-of-the-art strategies on two synthetic as well as ten real-world data sets for both regression and classification tasks.