AI 中文总结
本研究提出可微牛顿分子求解器（DINaMo），仅基于物理定律训练，无需模拟器生成的监督数据，在伦纳德-琼斯氩系统中成功复现分子轨迹的关键物理量，验证了轨迹无监督神经求解器的可行性。
AI 中文摘要
分子动力学（MD）模拟由显式运动方程支配，然而大多数用于加速或模拟MD的神经方法依赖模拟器生成的轨迹、力或能量进行训练。本研究探究能在多大程度上从支配定律中恢复具有物理意义的分子轨迹。我们提出可微牛顿分子求解器（DINaMo），这是一种物理信息神经框架，将分子轨迹表示为时间的可微函数，仅通过牛顿动力学、守恒定律以及给定平衡初始状态下的解析相互作用势进行训练。与先前的物理信息MD公式不同，DINaMo不使用模拟器生成的轨迹、力、速度或能量作为监督目标。在伦纳德-琼斯氩系统中，学习到的轨迹能复现短时间尺度的坐标、能量和结构可观测量，包括在更大、更致密的类液体环境中，还能恢复径向分布函数。尽管目前局限于短时间范围，结果表明具有物理意义的分子轨迹可直接从纯物理监督中生成，为分子动力学的轨迹无监督神经求解器提供了可行性支持。
英文摘要
Molecular dynamics (MD) simulations are governed by explicit equations of motion, yet most neural approaches that accelerate or emulate MD rely on simulator-generated trajectories, forces, or energies for training. In this work we ask to what extent can physically meaningful molecular trajectories be recovered from the governing laws. We introduce the Differentiable Newtonian Molecular Solver (DINaMo), a physics-informed neural framework that represents molecular trajectories as differentiable functions of time and is trained exclusively through Newtonian dynamics, conservation laws, and analytic interaction potentials on a given equilibrated initial state. Unlike prior physics-informed MD formulations, DINaMo uses no simulator-generated trajectories, forces, velocities, or energies as supervisory targets. In Lennard--Jones argon systems, the learned trajectories reproduce short-time coordinate, energy, and structural observables, including in a larger and denser liquid-like setting where the radial distribution function is recovered. Although currently limited to short temporal horizons, the results indicate that physically meaningful molecular trajectories can emerge directly from physics-only supervision, supporting the feasibility of trajectory-unsupervised neural solvers for molecular dynamics.
Comments26 pages, 6 figures, 5 tables
