Expected Improvements for the Asynchronous Parallel Global Optimization of Expensive Functions: Potentials and Challenges

Januševskis, Jānis; Riche, Rodolphe Le; Ginsbourger, David; Girdziušas, Ramūnas

doi:10.1007/978-3-642-34413-8_37

Cited by 45 publications

(32 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The multi-point Expected Improvement has received some attention recently, see for instance [22,21,26,27,18], where the criterion is computed using Monte-Carlo simulations. It has been calculated in closed form for q = 2 in [23] and extended for any q in [10].…”

Section: Targeted Bayesian Multi-objective Optimization By Batchmentioning

confidence: 99%

Targeting solutions in Bayesian multi-objective optimization: sequential and batch versions

Gaudrie

Riche

Picheny³

et al. 2019

Ann Math Artif Intell

Self Cite

View full text Add to dashboard Cite

Multi-objective optimization aims at finding trade-off solutions to conflicting objectives. These constitute the Pareto optimal set. In the context of expensive-to-evaluate functions, it is impossible and often non-informative to look for the entire set. As an end-user would typically prefer a certain part of the objective space, we modify the Bayesian multi-objective optimization algorithm which uses Gaussian Processes and works by maximizing the Expected Hypervolume Improvement, to focus the search in the preferred region. The cumulated effects of the Gaussian Processes and the targeting strategy lead to a particularly efficient convergence to the desired part of the Pareto set. To take advantage of parallel computing, a multi-point extension of the targeting criterion is proposed and analyzed.

show abstract

Section: Targeted Bayesian Multi-objective Optimization By Batchmentioning

confidence: 99%

Targeting solutions in Bayesian multi-objective optimization: sequential and batch versions

Gaudrie

Riche

Picheny³

et al. 2019

Ann Math Artif Intell

Self Cite

View full text Add to dashboard Cite

show abstract

“…Ginsbourger et al (2007) also proposed an alternate algorithm, dubbed "Constant Liar" that uses a constant value (such as the minimum, mean, or maximum of the function values). Janusevskis et al (2012) tackled directly the q-EI problem by Monte Carlo simulation, instead of using the Kriging Believer or Constant Liar alternatives, but that came with considerable cost. Frazier (2012) proposed a q-EI stochastic gradient approach that avoided the explicit calculation of q-EI.…”

Section: Ego -Expected Improvementmentioning

confidence: 99%

Parallel surrogate-assisted global optimization with expensive functions – a survey

Haftka

Villanueva

Chaudhuri

2016

Struct Multidisc Optim

205

View full text Add to dashboard Cite

Surrogate assisted global optimization is gaining popularity. Similarly, modern advances in computing power increasingly rely on parallelization rather than faster processors. This paper examines some of the methods used to take advantage of parallelization in surrogate based global optimization. A key issue focused on in this review is how different algorithms balance exploration and exploitation. Most of the papers surveyed are adaptive samplers that employ Gaussian Process or Kriging surrogates. These allow sophisticated approaches for balancing exploration and exploitation and even allow to develop algorithms with calculable rate of convergence as function of the number of parallel processors. In addition to optimization based on adaptive sampling, surrogate assisted parallel evolutionary algorithms are also surveyed. Beyond a review of the present state of the art, the paper also argues that methods that provide easy parallelization, like multiple parallel runs, or methods that rely on population of designs for diversity deserve more attention.

show abstract

“…Exploration happens when areas of high uncertainty are evaluated, and exploitation when areas with high expected utility are evaluated. Parallel asynchronous Expected Improvement (EI) can be used as an acquisition function, E[I (µ,λ) (X λ )] [10]. At any given time µ nodes are busy and λ nodes are idle.…”

Section: Auto-transfer Algorithmmentioning

confidence: 99%

Knowledge Transfer in Automatic Optimisation of Reconfigurable Designs

Kurek

Deisenroth

Luk

et al. 2016

2016 IEEE 24th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)

View full text Add to dashboard Cite

Abstract-This paper presents a novel approach for automatic optimisation of reconfigurable design parameters based on knowledge transfer. The key idea is to make use of insights derived from optimising related designs to benefit future optimisations. We show how to use designs targeting one device to speed up optimisation of another device. The proposed approach is evaluated based on various applications including computational finance and seismic imaging. It is capable of achieving up to 35% reduction in optimisation time in producing designs with similar performance, compared to alternative optimisation methods.

show abstract

Expected Improvements for the Asynchronous Parallel Global Optimization of Expensive Functions: Potentials and Challenges

Cited by 45 publications

References 8 publications

Targeting solutions in Bayesian multi-objective optimization: sequential and batch versions

Targeting solutions in Bayesian multi-objective optimization: sequential and batch versions

Parallel surrogate-assisted global optimization with expensive functions – a survey

Knowledge Transfer in Automatic Optimisation of Reconfigurable Designs

Contact Info

Product

Resources

About