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: h21c138f3d84646f1
Venue: VLDB
Year: 2026
Pagerank: 4.9793485e-05
Overall Rank: 10,722 | 27.92%
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
975	Democratizing Data Science through Interactive Curation of ML Pipelines	2019	SIGMOD	0.00012750518
1,568	HELIX: Holistic Optimization for Accelerating Iterative Machine Learning	2019	VLDB	0.0001021302
1,691	MISTIQUE: A System to Store and Query Model Intermediates for Model Diagnosis	2018	SIGMOD	9.8570722e-05
3,013	Saga: A Platform for Continuous Construction and Serving of Knowledge At Scale	2022	SIGMOD	7.7541678e-05
3,067	Oracle AutoML: A Fast and Predictive AutoML Pipeline	2020	VLDB	7.6876681e-05
4,334	LIMA: Fine-grained Lineage Tracing and Reuse in Machine Learning Systems	2021	SIGMOD	6.6569314e-05
5,773	Optimizing Data Pipelines for Machine Learning in Feature Stores	2023	VLDB	5.9987664e-05
5,792	Optimizing Machine Learning Workloads in Collaborative Environments	2020	SIGMOD	5.99349e-05
5,933	Doing More with Less: Characterizing Dataset Downsampling for AutoML	2021	VLDB	5.942188e-05
7,275	SubStrat: A Subset-Based Optimization Strategy for Faster AutoML	2023	VLDB	5.5679786e-05
8,310	DORIAN in action: Assisted Design of Data Science Pipelines	2022	VLDB	5.3581954e-05
9,117	Modyn: Data-Centric Machine Learning Pipeline Orchestration	2025	SIGMOD	5.2263399e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	7,613	The Machine Learning Bazaar: Harnessing the ML Ecosystem for Effective System Development	2020	SIGMOD
2	7,275	SubStrat: A Subset-Based Optimization Strategy for Faster AutoML	2023	VLDB
3	2,187	Production Machine Learning Pipelines: Empirical Analysis and Optimization Opportunities	2021	SIGMOD
4	4,524	DiffPrep: Differentiable Data Preprocessing Pipeline Search for Learning over Tabular Data	2023	SIGMOD
5	5,933	Doing More with Less: Characterizing Dataset Downsampling for AutoML	2021	VLDB
6	5,848	HAIPipe: Combining Human-generated and Machine-generated Pipelines for Data Preparation	2023	SIGMOD
7	6,921	CtxPipe: Context-aware Data Preparation Pipeline Construction for Machine Learning	2024	SIGMOD
8	10,990	MAPPipe: A System for Bridging Efficiency and Quality in Data Preprocessing via Knowledge-Augmented Structural Pruning	2026	VLDB
9	3,067	Oracle AutoML: A Fast and Predictive AutoML Pipeline	2020	VLDB
10	6,028	A Scalable AutoML Approach Based on Graph Neural Networks	2022	VLDB