Database Paper Browser

Back to papers

mlwhatif: What If You Could Stop Re-Implementing Your Machine Learning Pipeline Analyses Over and Over?

Summary: Declarative framework for data-centric ML what-if analyses that lets users specify pipeline perturbations and automatically generates, optimizes, and executes the required pipeline variants instead of reimplementing analyses. Demonstrates planner-level optimizations for robustness, data-cleaning and preprocessing/fairness studies across diverse pipelines; open-source mlwhatif library available. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID
13250
Venue
VLDB
Year
2023
Pagerank
4.5823351e-05
Overall Rank
8,114 | 43.56%
DOI
10.14778/3611540.3611606

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank Citing Paper Year Venue Pagerank
9,365 Falcon: Fair Active Learning using Multi-armed Bandits 2024 VLDB 4.3502315e-05
10,392 Shapley Value Estimation Based on Differential Matrix 2025 SIGMOD 4.1945683e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 5 of 5 cited papers.

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

Previous Page 1 / 1 Next

Semantically Similar Papers