Hybrid differentiable neural modeling
(Integrate numerical PDEs with deep neural networks via differentiable programming for Physics-informed hybrid DL architecture)

GPU-accelerated, differentiable physics
(Developing GPU-optimized differentiable solvers for flow, FSI, heat transfer and other complex multi-scale, multi-physics systems )

Generative AI for forward/inverse modeling
(TBD)

AI-Enabled Control for Physical systems
(Leveraging RL and data-driven methods to develop intelligent control strategies for complex dynamical systems, including flow/FSI control, and thermal management.)

Complex Geometry modeling & morphing
(TBD)

Dimension and model reduction
(TBD)

Data-driven surrogate modeling
(TBD)

Data assimilation & active learning
(TBD)

Bayesian uncertainty quantification
(TBD)

Wall-bounded turbulence
(TBD)

Fluid-structure interaction
(TBD)

Patient-specific image-based modeling
(TBD)

Aerodynamics & hydrodynamics
(TBD)

Personalized cardiovascular biomedicine
(TBD)

Biological and biomedical systems
(TBD)

Multiscale thermal management of nanoelectronics
(TBD)

Electrochemical machining & manufacturing
(TBD)

Fiber-reinforced composite manufacturing
(TBD)