{"id":"arxiv/2607.15291","url":"https://arxiv.org/abs/2607.15291","tags":["math.NA","cs.LG"],"score":0,"title":"A Physics-Informed Neural Network with a Modified Lorentzian Activation for Nonlocal Gradient-Flow Equations in Dynamic Density Functional Theory","domain":"math.NA","source":"arxiv","authors":["Dimitrios Gourzoulidis","Soumaya Elkantassi","Serafim Kalliadasis"],"reviews":0,"subject":"math.NA","summary":"We develop a physics-informed neural network (PINN) framework for nonlocal partial differential equations arising in dynamic density functional theory (DDFT). Such equations are challenging for standard PINN methods because they involve nonlinearities, nonlocal interaction terms, and an underlying gradient-flow structure, often leading to slow convergence and difficult optimization. We adapt the PINN methodology to DDFT gradient-flow equations and introduce two computational components: a modified Lorentzian activation function that behaves approximately linearly for small inputs and decays…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-24","reactions":{}}