基于分数的神经常微分方程用于求解反应扩散方程
Score-based neural ordinary differential equations for reaction-diffusions
把扩散模型里常见的分数驱动输运扩展到了会改变质量的反应扩散问题,每个粒子带位置和权重,还配了相对熵层面的误差控制理论。做生成建模、科学计算或粒子方法数值求解的研究者可以看看,实验涵盖 OU、Fisher–KPP 和 Wasserstein–Fisher–Rao 三类方程。
arXiv 论文 2611.18467 提出一种求解反应扩散方程的拉格朗日数值方法。传统 probability-flow 公式用分数函数驱动的速度场确定性输运粒子,通常会守恒总质量,无法处理有源项的反应过程。该方法让每个粒子同时携带位置和物理权重,特征 ODE 把源项纳入对数密度及其梯度的演化,粒子权重负责追踪质量变化。
在训练机制上,作者用学习到的速度场加权重粒子训练来逼近这些动力学。对于非增反应率,论文给出相对熵不等式,支持把质量归一化的自洽残差作为训练目标,并在增长率有界时控制诱导密度的误差。算法把时间区间切成短窗口,每个窗口内用冻结的速度场生成粒子数据,回归网络速度;粒子从随周期自适应的初始分布采样,损失中部分按预测未来质量加权,因为反应会使初始质量很小的区域在后期变得重要。
数值实验覆盖三类问题:带线性源项的 Ornstein–Uhlenbeck 方程、Fisher–KPP 方程以及 Wasserstein–Fisher–Rao 梯度流。对做生成建模与科学计算的读者来说,这篇把分数模型与反应-输运结构结合的工作提供了一套带理论误差控制的训练框架。
Score-based neural ordinary differential equations for reaction-diffusions
A diffusion equation can be rewritten as a transport equation driven by the score function of the evolving density. Conventional probability-flow formulations, which move particles deterministically with this score-driven velocity, typically conserve total mass. In this work, we develop a Lagrangian method for reaction-diffusion equations in which each particle carries a position and a physical weight. The characteristic ODEs incorporate source terms into the evolution of the log-density and its gradient, while particle weights track changes in mass. We combine this representation with a learned velocity field and weighted particle training to approximate these dynamics. For non-increasing reaction rates, a relative-entropy inequality supports the mass-normalized self-consistency residual as the training objective. This objective can control the error of the induced density when the growth rate is bounded. The algorithm divides the time interval into short windows. In each window, a frozen velocity generates the particle data for a regression of the network velocity. We draw the particles from period-adapted initial sampling distributions and weight them in the loss partly by their predicted future mass, because the reaction can make regions with little initial mass important later. We numerically investigate the method on the Ornstein--Uhlenbeck (OU) equation with a linear source term, Fisher--KPP equations, and Wasserstein--Fisher--Rao gradient flows.
本条情报归属 AITOP「论文」核心追踪栏目。涉及核心实体与技术议题包括:score-based models、反应扩散方程、Fisher-KPP、神经ODE,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.LG」(@Jianguo Hou, Wuchen Li),经 AITOP 多维评分管线结构化提炼。