Answers are grounded in this paper and cite their sources.
The use of machine learning to estimate exposure effects introduces a dependence between the results of an empirical study and the value of the seed used to fix the pseudo-random number generator.
No takes yet. Share an insight, caveat, or question.
Naimi et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: