Fast machine-learning online optimization of ultra-cold-atom experiments.
Fast machine-learning online optimization of ultra-cold-atom experiments.
复制标题
DOI：
10.1038/srep25890
复制
发表时间：
2016-05-16
影响因子：
4.6
通讯作者：
Hush MR
中科院分区：
文献类型：
--
作者：
Wigley PB;Everitt PJ;van den Hengel A;Bastian JW;Sooriyabandara MA;McDonald GD;Hardman KS;Quinlivan CD;Manju P;Kuhn CC;Petersen IR;Luiten AN;Hope JJ;Robins NP;Hush MR
我们将基于机器学习的在线优化过程应用于玻色-爱因斯坦凝聚体（BEC）的生产。BEC通常具有指数蒸发坡道，对于两体s波相互作用的遍历动力学来说是最优的，没有其他损失率，但对于实际实验来说可能不是最优的。通过反复的机器控制科学实验和观察，我们的“学习者”发现了BEC生产的最佳蒸发坡道。与之前的工作相反，我们的学习器使用高斯过程来建立它控制的参数与BEC质量之间关系的统计模型。我们证明高斯过程机器学习能够发现一个斜坡，产生高质量的BECs，迭代次数比以前使用的在线优化技术少10倍。此外，我们还展示了开发的内部模型可用于确定在BEC创建中哪些参数是必要的，哪些不重要，从而深入了解系统的优化过程。
We apply an online optimization process based on machine learning to the production of Bose-Einstein condensates (BEC). BEC is typically created with an exponential evaporation ramp that is optimal for ergodic dynamics with two-body s-wave interactions and no other loss rates, but likely sub-optimal for real experiments. Through repeated machine-controlled scientific experimentation and observations our ‘learner’ discovers an optimal evaporation ramp for BEC production. In contrast to previous work, our learner uses a Gaussian process to develop a statistical model of the relationship between the parameters it controls and the quality of the BEC produced. We demonstrate that the Gaussian process machine learner is able to discover a ramp that produces high quality BECs in 10 times fewer iterations than a previously used online optimization technique. Furthermore, we show the internal model developed can be used to determine which parameters are essential in BEC creation and which are unimportant, providing insight into the optimization process of the system.