TY - JOUR
T1 - Adaptive Interface-PINNs (AdaI-PINNs): An Efficient Physics-Informed Neural Networks Framework for Interface Problems
AU - Roy , Sumanta
AU - Annavarapu , Chandrasekhar
AU - Roy , Pratanu
AU - Sarma , Antareep Kumar
JO - Communications in Computational Physics
VL - 3
SP - 603
EP - 622
PY - 2025
DA - 2025/03
SN - 37
DO - http://doi.org/10.4208/cicp.OA-2024-0131
UR - https://global-sci.org/intro/article_detail/cicp/23915.html
KW - PINN, I-PINNs, AdaI-PINNs, domain decomposition, interface problems, machine learning, physics-informed machine learning.
AB - <p style="text-align: justify;">We present an efficient physics-informed neural networks (PINNs) framework, termed Adaptive Interface-PINNs (AdaI-PINNs), to improve the modeling of
interface problems with discontinuous coefficients and/or interfacial jumps. This framework is an enhanced version of its predecessor, Interface PINNs or I-PINNs (Sarma et
al. [1]; https://doi.org/10.1016/j.cma.2024.117135), which involves domain decomposition and assignment of different predefined activation functions to the neural networks in each subdomain across a sharp interface, while keeping all other parameters
of the neural networks identical. In AdaI-PINNs, the activation functions vary solely in
their slopes, which are trained along with the other parameters of the neural networks.
This makes the AdaI-PINNs framework fully automated without requiring preset activation functions. Comparative studies on one-dimensional, two-dimensional, and
three-dimensional benchmark elliptic interface problems reveal that AdaI-PINNs outperform I-PINNs, reducing computational costs by 2-6 times while producing similar
or better accuracy.</p>