What is it about?
How to learn the signal from an environment with changing noise? Based all information we have, how to collect data in the next step to obtain the maximal information under heteroskedasticity? How to balance exploration and replication to collect data and learn?
Why is it important?
Sequential design methodology in stochastic simulation is new. Our method is efficient both computationally and enjoys nice statistical properties.
Read the Original
This page is a summary of: Replication or exploration? Sequential design for stochastic simulation experiments, Technometrics, May 2018, Taylor & Francis, DOI: 10.1080/00401706.2018.1469433.
You can read the full text:
The following have contributed to this page