论文标题
深度学习核质量和$α$衰减的半衰期
Deep learning on nuclear mass and $α$ decay half-lives
论文作者
论文摘要
核质量，结合能和$α$衰减的半衰期的Ab-Initio计算对于重核很棘手，因为许多人体量子模拟的尺寸诅咒是质子编号（$ \ mathrm {n} $）（$ \ mathrm {n} $）和中子数（$ \ satrm {z} $）的生长。我们利用了强大的非线性转化和深神经网络（DNN）预测核质量和$α$衰减的半衰期的特征表示能力。对于核结合能量预测问题，我们实现了标准偏差$σ= 0.263 $ MEV在2149核上10倍的交叉验证上。从质量回归DNN隐藏层的核的高维表示的单词向量有助于我们计算$α$衰减的半衰期。对于此任务，我们在100倍的350核上获得$σ= 0.797 $，$ log_ {10} t_ {1/2} $，$σ= 0.731 $ 486 nuclei上的$ log_ {10} t_ {10} t_ {1/2} $。我们还发现了物理先验，例如外壳结构，魔术数量和受限范围液滴模型启发的增强输入对于此小型数据回归任务很重要。
Ab-initio calculations of nuclear masses, the binding energy and the $α$ decay half-lives are intractable for heavy nucleus, because of the curse of dimensionality in many body quantum simulations as proton number($\mathrm{N}$) and neutron number($\mathrm{Z}$) grow. We take advantage of the powerful non-linear transformation and feature representation ability of deep neural network(DNN) to predict the nuclear masses and $α$ decay half-lives. For nuclear binding energy prediction problem we achieve standard deviation $σ=0.263$ MeV on 10-fold cross validation on 2149 nuclei. Word-vectors which are high dimensional representation of nuclei from the hidden layers of mass-regression DNN help us to calculate $α$ decay half-lives. For this task, we get $σ=0.797$ on 100 times 10-fold cross validation on 350 nuclei on $log_{10}T_{1/2}$ and $σ=0.731 $ on 486 nuclei. We also find physical a priori such as shell structure, magic numbers and augmented inputs inspired by Finite Range Droplet Model are important for this small data regression task.
