CAPS: Cost-Aware ML Pipeline Selection

Summary: CAPS adds cost-aware pipeline selection to AutoML, orthogonal to the underlying search strategy, via lightweight time/cost estimation. Models candidate pipelines as a directed hypergraph and solves a constrained prize-collecting subset problem with a greedy approximation, cutting waste up to 4x. (summarized by gpt-5.4-mini on May 27 2026)

Paper ID: 14477
Venue: VLDB
Year: 2026
Pagerank: 5.093636e-05
Overall Rank: 10,540 | 27.69%
DOI: 10.14778/3801059.3801060

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Antonios Kontaxakis (Universitat Politècnica de Catalunya; Université Libre de Bruxelles)
2. Dimitris Sacharidis (Université Libre de Bruxelles)
3. Alberto Abelló (Universitat Politècnica de Catalunya)
4. Sergi Nadal (Universitat Politècnica de Catalunya)
5. Alkis Simitsis (Athena Research Center)

BibTeX Citation

@article{kontaxakis_vldb26,
        title = {{CAPS: Cost-Aware ML Pipeline Selection}},
        author = {Kontaxakis, Antonios and Sacharidis, Dimitris and Abelló, Alberto and Nadal, Sergi and Simitsis, Alkis},
        journal = {PVLDB},
        series = {{VLDB} '26},
        volume = {19},
        number = {7},
        pages = {1427--1440},
        doi = {10.14778/3801059.3801060},
        url = {https://doi.org/10.14778/3801059.3801060},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 12 of 12 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
1,004	Democratizing Data Science through Interactive Curation of ML Pipelines	2019	SIGMOD	0.00012701932
1,569	HELIX: Holistic Optimization for Accelerating Iterative Machine Learning	2019	VLDB	0.00010335423
1,670	MISTIQUE: A System to Store and Query Model Intermediates for Model Diagnosis	2018	SIGMOD	0.00010045615
3,012	Oracle AutoML: A Fast and Predictive AutoML Pipeline	2020	VLDB	7.8519448e-05
3,304	Saga: A Platform for Continuous Construction and Serving of Knowledge At Scale	2022	SIGMOD	7.5404346e-05
4,240	LIMA: Fine-grained Lineage Tracing and Reuse in Machine Learning Systems	2021	SIGMOD	6.809685e-05
5,699	Optimizing Machine Learning Workloads in Collaborative Environments	2020	SIGMOD	6.1170243e-05
5,704	Optimizing Data Pipelines for Machine Learning in Feature Stores	2023	VLDB	6.1146371e-05
5,839	Doing More with Less: Characterizing Dataset Downsampling for AutoML	2021	VLDB	6.0699037e-05
7,127	SubStrat: A Subset-Based Optimization Strategy for Faster AutoML	2023	VLDB	5.6957765e-05
8,134	DORIAN in action: Assisted Design of Data Science Pipelines	2022	VLDB	5.4811783e-05
8,949	Modyn: Data-Centric Machine Learning Pipeline Orchestration	2025	SIGMOD	5.3462965e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	7,000	A Cost-based Optimizer for Gradient Descent Optimization	2017	SIGMOD
2	7,473	The Machine Learning Bazaar: Harnessing the ML Ecosystem for Effective System Development	2020	SIGMOD
3	7,127	SubStrat: A Subset-Based Optimization Strategy for Faster AutoML	2023	VLDB
4	2,657	Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities	2021	SIGMOD
5	5,291	DiffPrep: Differentiable Data Preprocessing Pipeline Search for Learning over Tabular Data	2023	SIGMOD
6	5,839	Doing More with Less: Characterizing Dataset Downsampling for AutoML	2021	VLDB
7	8,177	HAIPipe: Combining Human-generated and Machine-generated Pipelines for Data Preparation	2023	SIGMOD
8	8,914	CtxPipe: Context-aware Data Preparation Pipeline Construction for Machine Learning	2024	SIGMOD
9	3,012	Oracle AutoML: A Fast and Predictive AutoML Pipeline	2020	VLDB
10	5,901	A Scalable AutoML Approach Based on Graph Neural Networks	2022	VLDB