batch.mlb: Multi-Level Batching Heuristic

In mludkov/mlOSP: Machine Learning and Regression Monte Carlo Algorithms for Optimal Stopping

Multi-Level Batching Heuristic

Description

Calculates weights for batch size in MLB when called from osp.seq.batch.design

Usage

batch.mlb(objSd_at_x_optim, r_cand, nugget, gamma)

Arguments

objSd_at_x_optim	posterior standard deviation of the response at the selected new input
r_cand	candidate batch size vector
nugget	the noise variance to compute the ALC factor
gamma	threshold compared with sd

Value

list containing:

• roptim: new replication count

• gamma: new gamma variable

References

M. Ludkovski, X. Lyu (2020+) Adaptive Batching for Gaussian Process Surrogates with Application in Noisy Level Set Estimation, http://arxiv.org/abs/2003.08579

See Also

osp.seq.batch.design

mludkov/mlOSP documentation built on April 29, 2023, 7:56 p.m.