New research explores physics-informed neural networks and operators · 10 sources tracked
ByPulseAugur Editorial·[11 sources]·
Multiple research papers explore advancements in physics-informed neural networks (PINNs) and neural operators. One study investigates whether improved scores in machine learning models directly translate to better physical accuracy for flow simulations around an airfoil. Another paper questions the reliance on automatic differentiation in PINNs, arguing that numerical precision does not guarantee physical fidelity. Additional research introduces novel architectures and training methods, such as data-free operators for interface advection, training with noisy Monte Carlo estimates for particle transport, and preconditioned losses for improved conditioning in physics-informed training. Techniques like causal integral terms and gradient surgery are also proposed to enhance the performance and scalability of PINNs for complex evolution equations and multi-task optimization problems.
AI
IMPACT
Advances in physics-informed neural networks and neural operators could lead to more accurate and efficient simulations in scientific domains.
RANK_REASON
Multiple arXiv papers presenting novel research in physics-informed neural networks and neural operators.
arXiv:2610.00415v1 Announce Type: new Abstract: Machine-learning surrogates accelerate physical simulation, but lower prediction error need not coincide with lower error in physically relevant flow statistics. We examine this question for flow around a NACA4418 airfoil using pair…
arXiv:2609.33078v2 Announce Type: replace Abstract: Automatic differentiation (AD) lets neural networks compute derivatives of governing equations to machine precision, and this precision has made it the computational backbone of physics-informed machine learning. Yet exactness i…
arXiv cs.LG
TIER_1English(EN)·Muhammad Akbar Khan·
arXiv:2609.38195v1 Announce Type: new Abstract: Operators for interfacial problems are trained on reference solutions produced by the solver they are intended to replace. This work develops a data-free physics-informed neural operator for level-set interface advection, in which t…
arXiv cs.AI
TIER_1English(EN)·Yubo Cao, Xi Deng, Mengqi Xia, Vignesh Gopakumar, Ander Gray, Anima Anandkumar·
arXiv:2609.40090v1 Announce Type: new Abstract: Particle transport under multiple scattering is central to radiative transfer and plasma physics, yet high-fidelity Monte Carlo (MC) simulations must trace prohibitively many particles. Learning-based surrogates can amortize this co…
arXiv cs.LG
TIER_1English(EN)·Alexandre Caboussat, Anna Peruso·
arXiv:2501.10162v3 Announce Type: replace-cross Abstract: Optimal transportation of raw material from suppliers to customers is an issue arising in logistics that is addressed here with a continuous model relying on optimal transport theory. A physics informed neural network meth…
arXiv cs.LG
TIER_1English(EN)·Carlo Marcati, Christoph Schwab·
arXiv:2112.08125v3 Announce Type: replace-cross Abstract: We construct and analyze approximation rates of deep operator networks (ONets) between infinite-dimensional spaces that emulate with an exponential rate of convergence the coefficient-to-solution map of elliptic second-ord…
arXiv cs.AI
TIER_1English(EN)·Huiwen Zhang, Feng Ye, Chu Ma·
arXiv:2609.38023v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) embed governing equations into deep learning, but enforce them only through loss residuals, leaving highly oscillatory wave behavior to be discovered by optimization. As a result, methods tha…
arXiv cs.LG
TIER_1English(EN)·Shizheng Wen, Siddhartha Mishra, Marius Zeinhofer·
arXiv:2609.36216v1 Announce Type: new Abstract: Neural operators are typically trained in a supervised fashion, which requires a dataset to be generated with a classical solver. Training them physics-informed, i.e., purely from the governing equations, removes this large offline …
arXiv cs.LG
TIER_1English(EN)·Xiaodong Feng, Ziyu Sun, Tao Tang, Xiaoliang Wan, Tao Zhou·
arXiv:2609.36615v1 Announce Type: cross Abstract: Physics-informed neural networks (PINNs) solve partial differential equations (PDEs) by incorporating governing physical laws into the training loss. For evolution equations, however, their conventional pointwise space--time repre…
arXiv cs.LG
TIER_1English(EN)·Thomas Borsani, Giuseppe Di Fatta·
arXiv:2609.30966v1 Announce Type: new Abstract: Physics-Informed Neural Networks (PINNs) are trained by optimising a composite objective that combines data fitting with physics-based constraints, typically resulting in a highly imbalanced multi-task optimisation problem. Under th…
We introduce Distance-Residual Physics-Informed Neural Networks (DR-PINNs), a physics-informed learning framework for approximating solutions of ordinary and partial differential inclusions (DIs), governing laws in which a differential operator is constrained to lie in a set-valu…
