Efficiently Mitigating the Impact of Data Drift on Machine Learning Pipelines
Summary: Introduces Data Distributions with Low Accuracy (DDLA), using decision trees to localize serving-data regions where drift harms black-box model accuracy. Enables selective retraining only for harmful drift, reducing pipeline costs while preserving predictive quality. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Sijie Dong (Université Paris Cité)
- 2. Qitong Wang (Université Paris Cité)
- 3. Soror Sahri (Université Paris Cité)
- 4. Themis Palpanas (Université Paris Cité)
- 5. Divesh Srivastava (AT&T)
BibTeX Citation
@article{dong_vldb24,
title = {{Efficiently Mitigating the Impact of Data Drift on Machine Learning Pipelines}},
author = {Dong, Sijie and Wang, Qitong and Sahri, Soror and Palpanas, Themis and Srivastava, Divesh},
journal = {PVLDB},
series = {{VLDB} '24},
volume = {17},
number = {11},
pages = {3072--3081},
doi = {10.14778/3681954.3681984},
url = {https://doi.org/10.14778/3681954.3681984},
year = {2024}
}Incoming Citations (Sorted by Pagerank)
Showing 1 of 1 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,864 | PipeLens: Identifying Interventions for Resolving Malfunctioning Data Science Pipelines | 2026 | VLDB | 4.9769913e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 483 | ActiveClean: Interactive Data Cleaning For Statistical Modeling | 2016 | VLDB | 0.00017584249 |
| 878 | Interpretable and Informative Explanations of Outcomes | 2015 | VLDB | 0.00013296412 |
| 1,308 | Automating Large-Scale Data Quality Verification | 2018 | VLDB | 0.00011073863 |
| 1,848 | Nearest Neighbor Classifiers over Incomplete Information: From Certain Answers to Certain Predictions | 2021 | VLDB | 9.5075544e-05 |
| 2,189 | Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities | 2021 | SIGMOD | 8.8854572e-05 |
| 3,738 | Learning to Validate the Predictions of Black Box Classifiers on Unseen Data | 2020 | SIGMOD | 7.0609398e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 7,529 | DeltaBoost: Gradient Boosting Decision Trees with Efficient Machine Unlearning | 2023 | SIGMOD |
| 2 | 4,675 | Data Platform for Machine Learning | 2019 | SIGMOD |
| 3 | 7,544 | Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines | 2023 | SIGMOD |
| 4 | 11,753 | An Experimental Evaluation of Process Concept Drift Detection | 2023 | VLDB |
| 5 | 2,189 | Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities | 2021 | SIGMOD |
| 6 | 9,426 | Towards Observability for Production Machine Learning Pipelines | 2022 | VLDB |
| 7 | 11,827 | Towards Observability for Machine Learning Pipelines | 2022 | CIDR |
| 8 | 5,715 | Detect, Distill and Update: Learned DB Systems Facing Out of Distribution Data | 2023 | SIGMOD |
| 9 | 10,424 | Are Learned DBMS Components Robust to Workload Drift?: [Experiments & Analysis] | 2026 | SIGMOD |
| 10 | 3,738 | Learning to Validate the Predictions of Black Box Classifiers on Unseen Data | 2020 | SIGMOD |