EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Budgeted Two-Stage Pre-Validation for Retrieval-Grounded Machine Learning Pipelines: A Group-Held-Out Evaluation

Maya Jenkins, Thomas Baker, Theo Hernandez

Academic Journal of Applied Sciences · 01.11.2026 · 3 (2) · 43-50

DOI: 10.54097/m33p3g15 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Academic Journal of Applied Sciences, online erschienen am 01.11.2026. Verfasst von Maya Jenkins und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Retrieval-grounded machine-learning pipeline synthesis must reconcile code templates with package versions and resource budgets. We study a two-stage pre-validator adapted from concrete dependency-satisfiability and resource-prediction mechanisms: an inexpensive retrieval-profile margin ranks candidate manifests; a budgeted second stage invokes a gradient-boosted peak-memory predictor and applies bounded repairs. Actual execution of 48 public-dataset learning workflows supplies memory profiles, while eight controlled configuration mutations per workload define explicit local feasibility outcomes. Four group-disjoint test folds provide 384 held-out candidates. At an exact 50% second-stage invocation budget, retrieval-prioritized validation yields 372/384 feasible configurations (96.9%), matching exhaustive checking, versus 281/384 (73.2%) for one matched random schedule. Across 300 random schedules the mean is 74.2%. A 16-group paired bootstrap puts the advantage over that seeded random baseline at 19.8–27.1 percentage points. Uncertainty offsets and semantic novelty are not independently beneficial on this coarse-fault benchmark: point prediction and a simple memory-margin ranking tie the proposed method. Calibration also deteriorates on a shifted fold. These findings support auditable, low-cost screening for synthetic configuration faults, without treating local feasibility as actual cloud deployment success or deterministic template synthesis as LLM generation.

Zitieren

APA

Jenkins, M., Baker, T., & Hernandez, T. (2026). Budgeted Two-Stage Pre-Validation for Retrieval-Grounded Machine Learning Pipelines: A Group-Held-Out Evaluation. Academic Journal of Applied Sciences, 3(2), 43-50. https://doi.org/10.54097/m33p3g15

Vancouver

Jenkins M, Baker T, Hernandez T. Budgeted Two-Stage Pre-Validation for Retrieval-Grounded Machine Learning Pipelines: A Group-Held-Out Evaluation. Academic Journal of Applied Sciences. 2026;3(2):43-50. doi:10.54097/m33p3g15

BibTeX
@article{Jenkins2026,
  author = {Jenkins, Maya and Baker, Thomas and Hernandez, Theo},
  title = {Budgeted Two-Stage Pre-Validation for Retrieval-Grounded Machine Learning Pipelines: A Group-Held-Out Evaluation},
  journal = {Academic Journal of Applied Sciences},
  year = {2026},
  volume = {3},
  number = {2},
  pages = {43--50},
  doi = {10.54097/m33p3g15}
}
Kostenlos starten →