arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1703.05511stat.CO
收藏

寻找最优贝叶斯试验设计的诱导自然选择启发式算法

An Induced Natural Selection Heuristic for Finding Optimal Bayesian Experimental Designs

David J. Price, Nigel G. Bean, Joshua V. Ross, Jonathan Tuke

摘要:

贝叶斯最优试验设计具有巨大的潜力,为数据的收集提供信息,从而提高我们对各种过程的理解。然而,一个主要的障碍是难以评估的问题,大,或高维,设计空间的最佳设计。我们提出了一个有效的搜索启发式适合一般的优化问题,特别注重最优贝叶斯试验设计问题。该启发式算法在初始随机生成的输入值集合处评估目标(效用)函数。在算法的每一代,如果输入值对应的目标(效用)函数满足某些接受标准,则输入值被“接受”,并且新的输入在这些接受的点附近被采样。我们证明了新的算法,通过评估最佳的贝叶斯试验设计为以前考虑的死亡,药代动力学和逻辑斯谛回归模型。与当前的“金标准”方法进行比较,以证明所提出的算法作为中等规模设计问题(即,高达大约40维)。

英文摘要:

Bayesian optimal experimental design has immense potential to inform the collection of data so as to subsequently enhance our understanding of a variety of processes. However, a major impediment is the difficulty in evaluating optimal designs for problems with large, or high-dimensional, design spaces. We propose an efficient search heuristic suitable for general optimisation problems, with a particular focus on optimal Bayesian experimental design problems. The heuristic evaluates the objective (utility) function at an initial, randomly generated set of input values. At each generation of the algorithm, input values are "accepted" if their corresponding objective (utility) function satisfies some acceptance criteria, and new inputs are sampled about these accepted points. We demonstrate the new algorithm by evaluating the optimal Bayesian experimental designs for the previously considered death, pharmacokinetic and logistic regression models. Comparisons to the current "gold-standard" method are given to demonstrate the proposed algorithm as a computationally-efficient alternative for moderately-large design problems (i.e., up to approximately 40-dimensions).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑