44 citations · 71 across the 9 of their papers we have counts for
7 papers · 1 filter
Differentiable Chemistry in PINNs for Solving Parameterized and Stiff Reaction Systems
Miloš Babić, Franz M. Rohrhofer, Stefan Posch
From neural ODEs to continuous-time machine learning, differentiable solvers allow physics, optimization, and simulation to become trainable components within deep learning systems…
Stabilizing PINNs: A regularization scheme for PINN training to avoid unstable fixed points of dynamical systems
Milos Babic, Franz M. Rohrhofer, Bernhard C. Geiger
It was recently shown that the loss function used for training physics-informed neural networks (PINNs) exhibits local minima at solutions corresponding to fixed points of dynamica…
B-PL-PINN: Stabilizing PINN Training with Bayesian Pseudo Labeling
Kevin Innerebner, Franz M. Rohrhofer, Bernhard C. Geiger
Training physics-informed neural networks (PINNs) for forward problems often suffers from severe convergence issues, hindering the propagation of information from regions where the…
Approximating Families of Sharp Solutions to Fisher's Equation with Physics-Informed Neural Networks
Franz M. Rohrhofer, Stefan Posch, Clemens Gößnitzer +1
This paper employs physics-informed neural networks (PINNs) to solve Fisher's equation, a fundamental reaction-diffusion system with both simplicity and significance. The focus is…
Bringing Chemistry to Scale: Loss Weight Adjustment for Multivariate Regression in Deep Learning of Thermochemical Processes
Franz M. Rohrhofer, Stefan Posch, Clemens Gößnitzer +2
Flamelet models are widely used in computational fluid dynamics to simulate thermochemical processes in turbulent combustion. These models typically employ memory-expensive lookup…
On the Role of Fixed Points of Dynamical Systems in Training Physics-Informed Neural Networks
Franz M. Rohrhofer, Stefan Posch, Clemens Gößnitzer +1
This paper empirically studies commonly observed training difficulties of Physics-Informed Neural Networks (PINNs) on dynamical systems. Our results indicate that fixed points whic…
Data vs. Physics: The Apparent Pareto Front of Physics-Informed Neural Networks
Franz M. Rohrhofer, Stefan Posch, Clemens Gößnitzer +1
Physics-informed neural networks (PINNs) have emerged as a promising deep learning method, capable of solving forward and inverse problems governed by differential equations. Despi…