ABSTRACT
Model Guided Sampling Optimization (MGSO) is a novel expensive black-box optimization method based on a combination of ideas from Estimation of Distribution Algorithms and global optimization methods using Gaussian Processes. The algorithm is described and its implementation tested on three benchmark functions as a proof of concept.
- S. Geman and D. Geman. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-6(6):721--741, 1984. Google ScholarDigital Library
- N. Hansen, S. Finck, R. Ros, and A. Auger. Real-parameter black-box optimization benchmarking 2009: Noiseless functions definitions. Technical Report RR-6829, INRIA, 2009. Updated February 2010.Google Scholar
- P. Hennig and C. J. Schuler. Entropy search for information-efficient global optimization. J. Mach. Learn. Res., 13:1809--1837, 2012. Google ScholarDigital Library
- D. Jones. A taxonomy of global optimization methods based on response surfaces. Journal of Global Optimization, 21:345--383, 2001. Google ScholarDigital Library
- M. Pelikan, K. Sastry, and E. Cant--u-Paz. Scalable Optimization via Probabilistic Modeling, volume 33 of Studies in Computational Intelligence. Springer, 2006. Google ScholarDigital Library
- C. E. Rasmussen and C. K. I. Williams. Gaussian Processes for Machine Learning. MIT Press, 2006. Google ScholarDigital Library
Index Terms
- Model guided sampling optimization with gaussian processes for expensive black-box optimization
Recommendations
Black-box optimization benchmarking for noiseless function testbed using particle swarm optimization
GECCO '09: Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking PapersThis paper benchmarks the Particle Swarm Optimization (PSO) algorithm using the noise-free BBOB 2009 testbed.
Black-box optimization benchmarking for noiseless function testbed using an EDA and PSO hybrid
GECCO '09: Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking PapersThis paper benchmarks an Estimation of Distribution Algorithm (EDA) and Particle Swarm Optimizer (PSO) on noise-free BBOB 2009 testbed. The algorithm is referred to as EDA-PSO and further enhanced with correlation-triggered adaptive variance scaling.
Black-box optimization benchmarking for noiseless function testbed using PSO_bounds
GECCO '09: Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking PapersThis paper benchmarks the particle swarm optimizer with adaptive bounds algorithm (PSO Bounds) on the noisefree BBOB 2009 testbed. The algorithm is further augmented with a simple re-initialization mechanism that is invoked if the bounds tend to ...
Comments