Machine learning in the string landscape
Machine learning in the string landscape
复制标题
DOI：
10.1007/jhep09(2017)157
复制
发表时间：
2017-09-28
影响因子：
5.4
通讯作者：
Nelson, Brent D.
中科院分区：
文献类型：
--
作者：
Carifio, Jonathan;Halverson, James;Nelson, Brent D.
我们利用机器学习来研究字符串景观。深度数据潜水和猜想生成被提出作为在景观中利用机器学习的有用框架，并给出了每个框架的示例。决策树精确地预测了由自反多面体产生的弱Fano复曲面三重数，每一个都确定了一个光滑的F-理论紧化，线性回归生成了一个先前证明过的关于4/3 x 2.96 x 10（755）个F-理论紧化系综中规范群秩的猜想。逻辑回归产生了一个新的猜想，当E-6出现在F-理论紧化的大系综，然后严格证明。这个结果可能与集合中可见扇区的出现有关。通过猜想生成，机器学习不仅对数字有用，而且对严格的结果也有用。
We utilize machine learning to study the string landscape. Deep data dives and conjecture generation are proposed as useful frameworks for utilizing machine learning in the landscape, and examples of each are presented. A decision tree accurately predicts the number of weak Fano toric threefolds arising from reflexive polytopes, each of which determines a smooth F-theory compactification, and linear regression generates a previously proven conjecture for the gauge group rank in an ensemble of 4/3 x 2.96 x 10(755) F-theory compactifications. Logistic regression generates a new conjecture for when E-6 arises in the large ensemble of F-theory compactifications, which is then rigorously proven. This result may be relevant for the appearance of visible sectors in the ensemble. Through conjecture generation, machine learning is useful not only for numerics, but also for rigorous results.