Differentiating the Multipoint Expected Improvement for Optimal Batch Design

This work deals with parallel optimization of expensive objective functions which are modelled as sample realizations of Gaussian processes. The study is formalized as a Bayesian optimization problem, or continuous multi-armed bandit problem, where a batch of q > 0 arms is pulled in parallel at each iteration. Several algorithms have been developed for choosing batches by trading off exploitation and exploration. As of today, the maximum Expected Improvement (EI) and Upper Confidence Bound (UCB) selection rules appear as the most prominent approaches for batch selection. Here, we build upon recent work on the multipoint Expected Improvement criterion, for which an analytic expansion relying on Tallis’ formula was recently established. The computational burden of this selection rule being still an issue in application, we derive a closed-form expression for the gradient of the multipoint Expected Improvement, which aims at facilitating its maximization using gradient-based ascent algorithms. Substantial computational savings are shown in application. In addition, our algorithms are tested numerically and compared to state-of-the-art UCB-based batchsequential algorithms. Combining starting designs relying on UCB with gradient-based EI local optimization finally appears as a sound option for batch design in distributed Gaussian Process optimization.

Using crop simulation to generate genotype by environment interaction effects for sorghum in water-limited environments

Multi-environment trials (METs) used to evaluate breeding lines vary in the number of years that they sample. We used a cropping systems model to simulate the target population of environments (TPE) for 6 locations over 108 years for 54 'near-isolines' of sorghum in north-eastern Australia. For a single reference genotype, each of 547 trials was clustered into 1 of 3 'drought environment types' (DETs) based on a seasonal water stress index. Within sequential METs of 2 years duration, the frequencies of these drought patterns often differed substantially from those derived for the entire TPE. This was reflected in variation in the mean yield of the reference genotype. For the TPE and for 2-year METs, restricted maximum likelihood methods were used to estimate components of genotypic and genotype by environment variance. These also varied substantially, although not in direct correlation with frequency of occurrence of different DETs over a 2-year period. Combined analysis over different numbers of seasons demonstrated the expected improvement in the correlation between MET estimates of genotype performance and the overall genotype averages as the number of seasons in the MET was increased.

The capital gains from trade are not enough: Evidence from the environmental accounts of Venezuela and Mexico

In principle, a country can not endure negative genuine savings for longperiods of time without experiencing declining consumption. Nevertheless,theoreticians envisage two alternatives to explain how an exporter ofnon-renewable natural resources could experience permanent negativegenuine savings and still ensure sustainability. The first one allegesthat the capital gains arising from the expected improvement in theterms of trade would suffice to compensate for the negative savings ofthe resource exporter. The second alternative points at technologicalchange as a way to avoid economic collapse. This paper uses the dataof Venezuela and Mexico to empirically test the first of these twohypotheses. The results presented here prove that the terms oftrade do not suffice to compensate the depletion of oil reservesin these two open economies.

Distance-based kriging relying on proxy simulations for inverse conditioning

Let us consider a large set of candidate parameter fields, such as hydraulic conductivity maps, on which we can run an accurate forward flow and transport simulation. We address the issue of rapidly identifying a subset of candidates whose response best match a reference response curve. In order to keep the number of calls to the accurate flow simulator computationally tractable, a recent distance-based approach relying on fast proxy simulations is revisited, and turned into a non-stationary kriging method where the covariance kernel is obtained by combining a classical kernel with the proxy. Once the accurate simulator has been run for an initial subset of parameter fields and a kriging metamodel has been inferred, the predictive distributions of misfits for the remaining parameter fields can be used as a guide to select candidate parameter fields in a sequential way. The proposed algorithm, Proxy-based Kriging for Sequential Inversion (ProKSI), relies on a variant of the Expected Improvement, a popular criterion for kriging-based global optimization. A statistical benchmark of ProKSI’s performances illustrates the efficiency and the robustness of the approach when using different kinds of proxies.

Quantile-Based Optimization of Noisy Computer Experiments with Tunable Precision

This article addresses the issue of kriging-based optimization of stochastic simulators. Many of these simulators depend on factors that tune the level of precision of the response, the gain in accuracy being at a price of computational time. The contribution of this work is two-fold: first, we propose a quantile-based criterion for the sequential design of experiments, in the fashion of the classical expected improvement criterion, which allows an elegant treatment of heterogeneous response precisions. Second, we present a procedure for the allocation of the computational time given to each measurement, allowing a better distribution of the computational effort and increased efficiency. Finally, the optimization method is applied to an original application in nuclear criticality safety. This article has supplementary material available online. The proposed criterion is available in the R package DiceOptim.

Global Optimization with Sparse and Local Gaussian Process Models

We present a novel surrogate model-based global optimization framework allowing a large number of function evaluations. The method, called SpLEGO, is based on a multi-scale expected improvement (EI) framework relying on both sparse and local Gaussian process (GP) models. First, a bi-objective approach relying on a global sparse GP model is used to determine potential next sampling regions. Local GP models are then constructed within each selected region. The method subsequently employs the standard expected improvement criterion to deal with the exploration-exploitation trade-off within selected local models, leading to a decision on where to perform the next function evaluation(s). The potential of our approach is demonstrated using the so-called Sparse Pseudo-input GP as a global model. The algorithm is tested on four benchmark problems, whose number of starting points ranges from 102 to 104. Our results show that SpLEGO is effective and capable of solving problems with large number of starting points, and it even provides significant advantages when compared with state-of-the-art EI algorithms.

Surrogate-based analysis and optimization for the design of heat sinks with jet impingement

Heat sinks are widely used for cooling electronic devices and systems. Their thermal performance is usually determined by the material, shape, and size of the heat sink. With the assistance of computational fluid dynamics (CFD) and surrogate-based optimization, heat sinks can be designed and optimized to achieve a high level of performance. In this paper, the design and optimization of a plate-fin-type heat sink cooled by impingement jet is presented. The flow and thermal fields are simulated using the CFD simulation; the thermal resistance of the heat sink is then estimated. A Kriging surrogate model is developed to approximate the objective function (thermal resistance) as a function of design variables. Surrogate-based optimization is implemented by adaptively adding infill points based on an integrated strategy of the minimum value, the maximum mean square error approach, and the expected improvement approaches. The results show the influence of design variables on the thermal resistance and give the optimal heat sink with lowest thermal resistance for given jet impingement conditions.

Online Supplement to ‘Myopic Allocation Policy with Asymptotically Optimal Sampling Rate’

This document is the Online Supplement to ‘Myopic Allocation Policy with Asymptotically Optimal Sampling Rate,’ to be published in the IEEE Transactions of Automatic Control in 2017.

Satisfaction with the healthcare provided of women who had undergone a mastectomy

Abstract Introduction: Breast cancer is the most common form of cancer among women worldwide and, therefore, deserves the highest attention and assistance from medical services. Considering patients’ satisfaction as an indication of healthcare quality, women who have undergone a mastectomy will assess the medical care received. This assessment will be based on what is expected from that medical care and on the expected improvement of her health condition. . Objective: To determine the level of satisfaction of women who have undergone a mastectomy with the medical care provided by nurses, doctors and by the way hospital services are organized. Design: A descriptive and cross-sectional study, developed in Portugal. Participants: A non-probabilistic sample formed by 153 women who underwent a mastectomy with an average age of 55, married (67.3%), unemployed (56.2%), living in a rural area (71.2%) and living on minimum wage (54.9%). Measurement instrument: European Organization for Research and Treatment of Cancer (EORTC) IN-PATSAT32 questionnaire. Results: 113 (73.85%) of the 153 women are satisied with the medical care provided and 40 (26.14%) of them show their lack of satisfaction. A highly signiicant percentage of women (49.01%) feel fairly satisied with the medical care provided by nurses and with the way services are organized (37.9). On the other hand (37.9%) show their dissatisfaction towards doctors. The family network proved to be a predictor of the satisfaction with doctors (β = 0.163; P = .044) and the period of hospitalization predicts the satisfaction with the organization (β = 0.171; P = .011). Both predictors will be useful to explain the 3% variability in patients’ satisfaction. Conclusions: Monitoring the satisfaction with the medical care received is a fundamental strategy to promote the well-being of women who underwent a mastectomy.

Stability improvement of the fatty acid binding protein Sm14 from S. mansoni by Cys replacement: Structural and functional characterization of a vaccine candidate

The Schistosoma mansoni fatty acid binding protein (FABP), SmA, is a vaccine candidate against, S. mansoni and F hepatica. Previously, we demonstrated the importance of a correct fold to achieve protection in immunized animals after cercariae challenge [[10]. C.R.R. Ramos, R.C.R. Figueredo, T.A. Pertinhez, M.M. Vilar, A.L.T.O. Nascimento, M. Tendler, I. Raw, A. Spisni, P.L. Ho, Gene structure and M20T polymorphism of the Schistosoma mansoni Sm14 fatty acid-binding protein: structural, functional and immunoprotection analysis. J. Biol. Chem. 278 (2003) 12745-12751]. Here we show that the reduction of vaccine efficacy over time is due to protein dimerization and subsequent aggregation. We produced the mutants Sm14-M20(C62S) and Sm14M20(C62V) that, as expected, did not dimerize in SDS-PAGE. Molecular dynamics calculations and unfolding experiments highlighted a higher structural stability of these mutants with respect to the wild-type. In addition, we found that the mutated proteins, after thermal denaturation, refolded to their active native molecular architecture as proved by the recovery of the fatty acid binding ability. Sm14-M20(C62V) turned out to be the more stable form over time, providing the basis to determine the first 3D solution structure of a Sm14 protein in its apo-form. Overall, Sm14-M20(C62V) possesses an improved structural stability over time, an essential feature to preserve its immunization capability and, in experimentally immunized animals, it exhibits a protection effect against S. mansoni cercariae infections comparable to the one obtained with the wild-type protein. These facts indicate this protein as a good lead molecule for large-scale production and for developing an effective Sm14 based anti-helminthes vaccine. (C) 2008 Elsevier B.V. All rights reserved.

Friction Stir Processing : a new microstructure improvement technique

In the present work, microstructure improvement using FSP (Friction Stir Processing) is studied. In the first part of the work, the microstructure improvement of as-cast A356 is demonstrated. Some tensile tests were applied to check the increase in ductility. However, the expected results couldn’t be achieved. In the second part, the microstructure improvement of a fusion weld in 1050 aluminium alloy is presented. Hardness tests were carried out to prove the mechanical propertyimprovements. In the third and last part, the microstructure improvement of 1050 aluminium alloy is achieved. A discussion of the mechanical property improvements induced by FSP is made. The influence of tool traverse speed on microstructure and mechanical properties is also discussed. Hardness tests and recrystallization theory enabled us to find out such influence

Improvement of saliva production in ENT cancer patients treated by IMRT

Intensity modulated radiotherapy (IMRT) is a conformal radiotherapy that produces concave and irregular target volume dose distributions. IMRT has a potential to reduce the volume of healthy tissue irradiated to a high dose, but this often at the price of an increased volume of normal tissue irradiated to a low dose. Clinical benefits of IMRT are expected to be most pronounced at the body sites where sensitive normal tissues surround or are located next to a target with a complex 3D shape. The irradiation doses needed for tumor control are often markedly higher than the tolerance of the radiation sensitive structures such as the spinal cord, the optic nerves, the eyes, or the salivary glands in the treatment of head and neck cancer. Parotid gland salivary flow is markedly reduced following a cumulative dose of 30 50 Gy given with conventional fractionation and xerostomia may be prevented in most patients using a conformal parotid-sparing radiotherapy technique. However, in cohort studies where IMRT was compared with conventional and conformal radiotherapy techniques in the treatment of laryngeal or oropharyngeal carcinoma, the dosimetric advantage of IMRT translated into a reduction of late salivary toxicity with no apparent adverse impact on the tumor control. IMRT might reduce the radiation dose to the major salivary glands and the risk of permanent xerostomia without compromizing the likelihood for cure. Alternatively, IMRT might allow the target dose escalation at a given level of normal tissue damage. We describe here the clinical results on postirradiation salivary gland function in head and neck cancer patients treated with IMRT, and the technical aspects of IMRT applied. The results suggest that the major salivary gland function can be maintained with IMRT without a need to compromise the clinical target volume dose, or the locoregional control.

Resolving inconsistencies in utility measurement under risk: Tests of generalizations of expected utility

This paper explores biases in the elicitation of utilities under risk and the contribution that generalizations of expected utility can make to the resolution of these biases. We used five methods to measure utilities under risk and found clear violations of expected utility. Of the theories studies, prospect theory was most consistent with our data. The main improvement of prospect theory over expected utility was in comparisons between a riskless and a risky prospect(riskless-risk methods). We observed no improvement over expected utility in comparisons between two risky prospects (risk-risk methods). An explanation why we found no improvement of prospect theory over expected utility in risk-risk methods may be that there was less overweighting of small probabilities in our study than has commonly been observed.

Council Bluffs Interstate System Improvement Project Tier 2, Segment 3 Pottawatomie County, Iowa Environmental Assessment, March 2011

This Tier 2 Environmental Assessment (EA) presents the results of studies and analyses conducted to determine the potential impacts of proposed improvements in Segment 3 of the Council Bluffs Interstate System (CBIS) in the Council Bluffs metropolitan area. This document is tiered to the Tier 1 Draft and Final Environmental Impact Statements (EIS) that evaluated impacts of the overall CBIS Improvements Project, which includes five segments of independent utility1 This EA on Segment 3 of the Project is divided into the following sections: and encompasses 18 mainline miles of Interstate and 14 interchanges along Interstate 80 (I-80), Interstate 29 (I-29), and Interstate 480 (I-480). More information about the tiering process is found below under Project Background. • Section 1 provides background information on the Project and discusses the relationship between the earlier Tier 1 EIS and this Tier 2 EA. It also discusses the proposed action and the area studied, the purpose of the Project, and the need for the Project based on transportation problems that currently exist or are expected in the future. • Section 2, Alternatives, identifies the range of alternatives considered for Segment 3 to address the transportation problems identified in Section 1. It also identifies the alternatives retained for further study in this EA and the preferred Segment 3 alternative. • Section 3, Affected Environment and Environmental Consequences, describes the general environment for each resource affected by the proposed improvements. It also describes the potential environmental impacts of the Segment 3 Project and methods to avoid, minimize, and mitigate impacts. • Section 4, Disposition, lists the agencies and organizations that will receive copies of this EA and the locations at which this EA will be available for public review. • Section 5, Comments and Coordination, summarizes the agency coordination and public involvement efforts in conjunction with the Segment 3 Project. • Section 6, Conclusion and Recommendation, summarizes resource impacts. • Section 7, References, lists the sources cited in this EA. For Segment 3, the Federal Highway Administration (FHWA) and Iowa Department of Transportation (Iowa DOT) determined that an EA is the appropriate level of Tier 2 study to comply with the National Environmental Policy Act (NEPA) requirements. The primary purpose of an EA is to clearly establish the significance of a project’s environmental impacts. That analysis is included in this document.