Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces `fastml`, an R package designed to prevent preprocessing leakage in automated machine learning workflows by implementing guarded resampling, where preprocessing steps are re-estimated within each resample. This approach mitigates inflated performance metrics caused by data-dependent transformations estimated before resampling. The package offers features like grouped resampling, dependency auditing, and sandboxed execution, and its effectiveness is validated through simulations and benchmarks demonstrating reduced performance inflation and streamlined workflow orchestration compared to standard methods.
Leakage-aware machine learning is now a one-liner in R: `fastml` prevents inflated performance metrics by re-estimating preprocessing steps within each resample, making your AutoML pipelines more reliable.
Preprocessing leakage arises when scaling, imputation, or other data-dependent transformations are estimated before resampling, inflating apparent performance while remaining hard to detect. We present fastml, an R package that provides a single-call interface for leakage-aware machine learning through guarded resampling, where preprocessing is re-estimated inside each resample and applied to the corresponding assessment data. The package supports grouped and time-ordered resampling, blocks high-risk configurations, audits recipes for external dependencies, and includes sandboxed execution and integrated model explanation. We evaluate fastml with a Monte Carlo simulation contrasting global and fold-local normalization, a usability comparison with tidymodels under matched specifications, and survival benchmarks across datasets of different sizes. The simulation demonstrates that global preprocessing substantially inflates apparent performance relative to guarded resampling. fastml matched held-out performance obtained with tidymodels while reducing workflow orchestration, and it supported consistent benchmarking of multiple survival model classes through a unified interface.