论文标题
在嘈杂级别设置估算中使用应用程序代理的自适应批处理
Adaptive Batching for Gaussian Process Surrogates with Application in Noisy Level Set Estimation
论文作者
论文摘要
我们为随机实验的高斯工艺元模型开发自适应复制设计。自适应批处理是顺序设计启发式方法的自然扩展，随着学习响应特征的学习，浓缩物和元模块的高架上升，复制增长的好处。在学习平均模拟器响应水平集的问题中，我们开发了四种新型方案：多级批处理（MLB），棘轮批次（RB），自适应批次逐步逐步降低不确定性（荒谬），具有逐步分配（ADSA）的自适应设计和具有逐步设计的逐步设计和逐步设计（DDSA）。同时（MLB，RB和荒谬）或顺序（ADSA和DDSA）同时确定顺序设计输入和相应的重复数量。使用合成示例和定量融资中应用的插图（通过回归蒙特卡洛的价格定价）表明，自适应批处理会带来显着的计算加速，并且最小的建模保真度损失。
We develop adaptive replicated designs for Gaussian process metamodels of stochastic experiments. Adaptive batching is a natural extension of sequential design heuristics with the benefit of replication growing as response features are learned, inputs concentrate, and the metamodeling overhead rises. Motivated by the problem of learning the level set of the mean simulator response we develop four novel schemes: Multi-Level Batching (MLB), Ratchet Batching (RB), Adaptive Batched Stepwise Uncertainty Reduction (ABSUR), Adaptive Design with Stepwise Allocation (ADSA) and Deterministic Design with Stepwise Allocation (DDSA). Our algorithms simultaneously (MLB, RB and ABSUR) or sequentially (ADSA and DDSA) determine the sequential design inputs and the respective number of replicates. Illustrations using synthetic examples and an application in quantitative finance (Bermudan option pricing via Regression Monte Carlo) show that adaptive batching brings significant computational speed-ups with minimal loss of modeling fidelity.
