Global optimization of quantum dynamics with AlphaZero deep exploration
Global optimization of quantum dynamics with AlphaZero deep exploration
复制标题
DOI：
10.1038/s41534-019-0241-0
复制
发表时间：
2020-01-14
影响因子：
7.6
通讯作者：
Sherson, Jacob
中科院分区：
文献类型：
--
作者：
Dalgaard, Mogens;Motzoi, Felix;Sherson, Jacob
虽然已经开发了大量针对不同目标优化量子动力学的算法，但一个常见的限制是依赖于良好的初始猜测，要么是随机的，要么是基于启发式和直觉的。在这里，我们实现了Deepmind AlphaZero算法的tabula深度量子探索版本，以系统地避免这种限制。AlphaZero在引导树搜索中使用深度神经网络与深度前瞻相结合，这允许对量子参数景观进行预测性隐藏变量近似。为了强调可转移性，我们只使用一个通用的算法超参数集在三类控制问题上应用该算法并对其进行基准测试。与早期的方法相比，AlphaZero在好的解决方案集群的质量和数量上都取得了实质性的改进。它能够自发地学习解中意想不到的隐藏结构和全局对称性，甚至超越了人类的启发式。
While a large number of algorithms for optimizing quantum dynamics for different objectives have been developed, a common limitation is the reliance on good initial guesses, being either random or based on heuristics and intuitions. Here we implement a tabula rasa deep quantum exploration version of the Deepmind AlphaZero algorithm for systematically averting this limitation. AlphaZero employs a deep neural network in conjunction with deep lookahead in a guided tree search, which allows for predictive hidden-variable approximation of the quantum parameter landscape. To emphasize transferability, we apply and benchmark the algorithm on three classes of control problems using only a single common set of algorithmic hyperparameters. AlphaZero achieves substantial improvements in both the quality and quantity of good solution clusters compared to earlier methods. It is able to spontaneously learn unexpected hidden structure and global symmetry in the solutions, going beyond even human heuristics.