SEGOMOE: Super Efficient Global Optimization with Mixture of Experts
SEGOMOE is a Bayesian optimization toolbox developed by ONERA and ISAE-SUPAERO. For more information refer to:
Bartoli, N., Lefebvre, T., Dubreuil, S., Olivanti, R., Priem, R., Bons, N., Martins J.R.R.A & Morlier, J. (2019). Adaptive modeling strategy for constrained global optimization with application to aerodynamic wing design. Aerospace Science and Technology, (90) 85-102.
Priem, R., Bartoli, N., Diouane, Y., & Sgueglia, A. (2020). Upper trust bound feasibility criterion for mixed constrained Bayesian optimization with application to aircraft design. Aerospace Science and Technology, 105980.
Saves, P., Bartoli, N., Diouane, Y., Lefebvre, T., Morlier, J., David, C., ... & Defoort, S. (2021, July). Constrained Bayesian optimization over mixed categorical variables, with application to aircraft design. In Proceedings of the International Conference on Multidisciplinary Design Optimization of Aerospace Systems (AEROBEST 2021) (pp. 1-758).
Installation
SEGOMOE is not openly available.
Usage
SEGOMOE is interacted with through the SEGOMOEInterface class. This class has a state containing evaluated (and
failed) points, and requires a directory for results storage. The run_optimization function can be used to
run the DOE and infill search.
from sb_arch_opt.algo.segomoe_interface import SEGOMOEInterface
problem = ... # Subclass of ArchOptProblemBase
# Define folder to store results in
results_folder = ...
# Use Mixture of Experts: automatically identifies clusters in the design space
# with different best surrogates ("experts"). Can be more accurate, however
# also greatly increases the cost of finding new infill points.
use_moe = True
# Options passed to the Sego class and to model generation, respectively
sego_options = {}
model_options = {}
# Get the interface (will be initialized if the results folder has results)
interface = SEGOMOEInterface(problem, results_folder, n_init=100, n_infill=50,
use_moe=use_moe, sego_options=sego_options,
model_options=model_options)
# Initialize from other results if you want
interface.initialize_from_previous('path/to/other/results_folder')
# Run the optimization loop incl DOE
interface.run_optimization()
x = interface.x # (n, nx)
x_failed = interface.x_failed # (n_failed, nx)
f = interface.f # (n, nf)
g = interface.g # (n, ng)
pop = interface.pop # Population containing all design points
opt = interface.opt # Population containing optimal point(s)
pymoo API
It is also possible to use the pymoo API to run an optimization:
from pymoo.optimize import minimize
from sb_arch_opt.algo.segomoe_interface import SEGOMOEInterface, SEGOMOEAlgorithm
problem = ... # Subclass of ArchOptProblemBase
# Define folder to store results in
results_folder = ...
# Use Mixture of Experts: automatically identifies clusters in the design space
# with different best surrogates ("experts"). Can be more accurate, however
# also greatly increases the cost of finding new infill points.
use_moe = True
# Options passed to the Sego class and to model generation, respectively
sego_options = {}
model_options = {}
# Get the interface (will be initialized if the results folder has results)
interface = SEGOMOEInterface(problem, results_folder, n_init=100, n_infill=50,
use_moe=use_moe, sego_options=sego_options,
model_options=model_options)
# Define the pymoo Algorithm
algo = SEGOMOEAlgorithm(interface)
# Initialize from other results if you want
algo.initialize_from_previous_results(problem, results_folder='/optional/other/result/folder')
# Run the optimization
# Note: no need to give a termination, as that is already defined by the SEGOMOEInterface object (n_init + n_infill)
result = minimize(problem, algo, seed=42) # Remove seed in production