Project: New Architectures and Training Algorithms for SCIML
Our goals in this project concern new architectures and new algorithms for physics-informed neural networks, neural operators, and Bayesian models.
Representative publications:
- Luna, Braga-Neto, Raissi, Gildin (2026) SoL-DeepONet: Solver-In-The-Loop Deep Operator Networks for Parametric PDEs
- Chiu, Cheung, Braga-Neto, Lee, Li (2026) Free-RBF-KAN: Kolmogorov-Arnold Networks with Adaptive Radial Basis Functions for Efficient Function Learning.
- Chiu, Nambiar, Syed, Siegel, Braga-Neto (2025) In-Context Multi-Operator Learning with DeepOSets
- Iqbal, Abdulsamad, Cator, Braga-Neto, Särkkä (2024) Parallel-in-time probabilistic solutions for time-dependent nonlinear partial differential equations.
- McClenny and Braga-Neto (2023) Self-Adaptive Physics-Informed Neural Networks.
- Coutinho, Dall’Aqua, McClenny, Zhong, Braga-Neto, Gildin (2023) Physics-informed neural networks with adaptive localized artificial viscosity.
- Davi and Braga-Neto (2022): PSO-PINN: Physics-Informed Neural Networks Trained with Particle Swarm Optimization.
- Braga-Neto (2022): Characteristics-Informed Neural Networks for Forward and Inverse Hyperbolic Problems.
Project: Reservoir Simulation
(This is a pilot SciML Lab project proposed by E. Gildin and U. Braga-Neto)
This project explores new approaches to physics-based and data-driven models and combinations thereof that effectively embeds physical constraints, conservation laws and constitutive relations with data for robust upscaling, sensible time-stepping, accurate proxy models and efficient optimization schemes.
Representative publications:
- Zhang, Chiu, Braga-Neto and Gildin (2025) Physics-Informed Neural Networks for CO2 migration modeling in stratified saline aquifers: Applications in geological carbon sequestration.
- Zhang, Braga-Neto and Gildin (2024) Physics-informed neural networks for multiphase flow in porous media considering dual shocks and interphase solubility
Project: Thermonuclear Supernovae: A Deep Neural Network Approach to the Explosion Physics
(This is a pilot SciML Lab project proposed by J. Tao, L. Perez and U. Braga-Neto)
This project will lead to new approaches to Type Ia supernovae (SN~Ia) cosmology that can quantitatively incorporate the latest in observational and theoretical developments using SciML tools.
Representative publications:
- Chen, Braga-Neto, Wang, Kasen, Liu, Röpke, Zhong, and Jeffery (2025) SEDONA-GesaRaT: an AI-Accelerated Radiative Transfer Program for 3-D Supernova Simulations
- Chen, Jeffery, Zhong, McClenny, Braga-Neto and Lifan Wang (2022) Using physics informed neural networks for supernova radiative transfer simulation
Project: Nuclear Fusion
This project is a recent collaboration with the Institute of Fusion Sciences at UT-Austin, Virginia Tech, and VTT (Finland). The goal is to develop deep learning surrogates for fast simulation of plasma turbulence in fusion devices.
Representative publication:
- Chen, Poole, Farcas, Hatch and Braga-Neto (2026) Convolution Operator Network for Forward and Inverse Problems (FI-Conv): Application to Plasma Turbulence Simulations
Project: Microstructure Informatics
(This is a pilot SciML Lab project proposed by R. Arroyave and U. Braga-Neto)
The goal of this project is to develop novel scientific machine learning (SciML) algorithms and to apply them to highly coupled multi-physics thermodynamically consistent phase field models for microstructure evolution, in order to deliver a set of efficient and accurate physics-aware forward and inverse models of processing-microstructure relationships in nanostructured composite thermoelectric materials, together with reliable uncertainty quantification metrics to support decision making in the design of these materials.
Representative publication:
- Zhong, Liu, Arroyave and Braga-Neto (2024) Label propagation training schemes for physics-informed neural networks and Gaussian processes