用神经网络精确计算量子激发状态
David Pfau1,2, Simon Axelrod1,3,4, Halvard Sutterud2
1Google DeepMind, London N1C 4DJ, UK.
Science (New York, N.Y.)
|August 22, 2024
概括
我们开发了一个新的变量蒙特卡洛算法来计算没有自由参数的量子系统激发状态. 这种方法准确地确定分子的激发能量和振荡器强度,包括具有挑战性的双激发.
科学领域:
- 量子力学
- 计算化学
- 量子物理学
背景情况:
- 计算量子系统的激发状态在计算上具有挑战性.
- 传统方法通常需要自由参数或状态正交.
- 准确预测分子特性对于化学研究至关重要.
研究的目的:
- 介绍一个新的,无参数算法来估计激发状态.
- 允许任意可观测的计算,包括非对角的元素.
- 提高激发能量和振荡器强度预测的准确性.
主要方法:
- 使用扩展系统方法的变量蒙特卡洛 (VMC).
- 将激发状态计算转化为基本状态问题.
- 采用神经网络的方法,
主要成果:
- 精确回收各种分子的激发能量和振荡器强度.
- 成功计算了尺度分子的垂直激发能.
- 证明了处理具有挑战性的双重激发的能力.
结论:
- 提出的VMC算法为激发状态计算提供了一个强大的,无参数的方法.
- 这种技术在与先进的神经网络相结合时显示出高精度.
- 这种方法在原子,核和凝聚物质物理学中具有广泛的适用性.
相关概念视频
01:19The Role of Ion Channels in Neuronal Computation
3.2K
A postsynaptic neuron usually receives numerous impulses from several other presynaptic neurons. The axon hillock of the postsynaptic neuron integrates all these signals and determines the likelihood of firing an action potential.
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
Sometimes a single EPSP is strong enough to induce an action potential in the postsynaptic neuron. However, multiple presynaptic inputs must often create EPSPs around the same time for the postsynaptic neuron to be sufficiently depolarized to fire an action potential....
3.2K
01:23Atomic Nuclei: Nuclear Relaxation Processes
632
In the absence of an external magnetic field, nuclear spin states are degenerate and randomly oriented. When a magnetic field is applied, the spins begin to precess and orient themselves along (lower energy) or against (higher energy) the direction of the field. At equilibrium, a slight excess population of spins exists in the lower energy state. Because the direction of the magnetic field is fixed as the z-axis, the precessing magnetic moments are randomly oriented around the z-axis.
632