다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Statistics for Spatial Data.

Surrogate-based Analysis and Optimization

https://eprints.soton.ac.uk/65935/01/Forr_09.pdf

Recent advances in surrogate-based optimization

Surrogate-assisted, or meta-model based evolutionary computation uses efficient computational models, often known as surrogates or meta-models, for approximating the fitness function in evolutionary algorithms. Research on surrogate-assisted evolutionary computation began over a decade ago and has received considerably increasing interest in recent years. Very interestingly, surrogate-assisted evolutionary computation has found successful applications not only in solving computationally expensive single- or multi-objective optimization problems, but also in addressing dynamic optimization problems, constrained optimization problems and multi-modal optimization problems. This paper provides a concise overview of the history and recent developments in surrogate-assisted evolutionary computation and suggests a few future trends in this research area.

http://www.soft-computing.de/SECPublished.pdf

Surrogate-assisted evolutionary computation: Recent advances and future challenges

The custom in surrogate-based modeling of complex engineering problems is to fit one or more surrogate models and select the one surrogate model that performs best. In this paper, we extend the utility of an ensemble of surrogates to (1) identify regions of possible high errors at locations where predictions of surrogates widely differ, and (2) provide a more robust approximation approach. We explore the possibility of using the best surrogate or a weighted average surrogate model instead of individual surrogate models. The weights associated with each surrogate model are determined based on the errors in surrogates. We demonstrate the advantages of an ensemble of surrogates using analytical problems and one engineering problem. We show that for a single problem the choice of test surrogate can depend on the design of experiments.

Ensemble of surrogates

Objectives This study presents a reliable index inspired by the total variability concept of real analysis in mathematics, called average real variability (ARV), for the prognostic significance of blood pressure variability (BPV) overcoming the pitfalls of the commonly used standard deviation (SD). Background Recent studies have suggested that an increase in BPV is associated with an increase in subsequent cardiovascular events/complications. However, there are other studies where the cited association was not found or was lost in the presence of other well-known risk factors. An explanation for these apparently contradictory results may be the selection of the variability index used (SD). Methods Ambulatory blood pressure monitoring in 312 subjects aged > 55 years. Logistic regression models and survival methods were used to establish the prognostic significance of awake systolic BPV: in particular, (i) the performance of ARV versus SD, and (ii) the value of BPV relative to other well-known risk factors. Results The analyses using the ARV index show a statistically significant relative risk equal to 4.548 (P U 0.006) for the group with high BPV with respect to the low BPV group (reference level); in contrast, the corresponding relative risk associated to the SD index was not statistically significant. Furthermore, ARV exhibited a similar predictive value to systolic blood pressure. Conclusions The proposed ARV index is a more reliable representation of time series variability than SD and may be less sensitive to the relative low sampling frequency of the ambulatory blood pressure monitoring devices. The results suggest that ARV adds prognostic value to the ABPM and could prompt the use of therapeutic measures to control BPV. J Hypertens 23:505‐511 Q 2005 Lippincott Williams & Wilkins.

Nestor V. Queipo

Papers

Surrogate-based Analysis and Optimization

Ensemble of surrogates

A reliable index for the prognostic significance of blood pressure variability

Response surface approximation of Pareto optimal front in multi-objective optimization

An optimization methodology of alkaline-surfactant-polymer flooding processes using field scale numerical simulation and multiple surrogates