Skip to main content
. 2019 Jun 3;43(6):e12738. doi: 10.1111/cogs.12738

Figure 6.

Figure 6

Average model fit and confidence bounds for different point estimation methods as a function of the number of total CPU‐hours used. The shaded region indicates the area between the 25th and 75th percentiles. ABC PM is the ABC posterior mean; BO is Bayesian optimization. Each point is estimated using 10 independent experiments, except for parallel Nelder–Mead, which is simulated using the non‐parallel experiments.