Publications

Multi-resolution partial differential equations preserved learning framework for spatiotemporal dynamics

Published in Communications Physics, 2024

Embedding Finite Difference Operators into Neural Networks in a Multi-Resolution manner

[ Paper | Code ]

Predicting 3D soft tissue dynamics from 2D imaging using physics informed neural networks

Published in Communications Biology, 2023

[ Paper ]

Physics-informed Dyna-style model-based deep reinforcement learning for dynamic control

Published in Proceedings of the Royal Society A, 2021

Using PINN as a surrogate environment for fast, yet accurate environment interactions

[ Paper | Code ]

CoNFiLD: Conditional Neural Field Latent Diffusion Model Generating Spatiotemporal Turbulence

Published in Arxiv, 2024

This study introduces the Conditional Neural Field Latent Diffusion (CoNFiLD) model, a novel generative learning framework designed for rapid simulation of intricate spatiotemporal dynamics in chaotic and turbulent systems within three-dimensional irregular domains. Traditional eddy-resolved numerical simulations, despite offering detailed flow predictions, encounter significant limitations due to their extensive computational demands, restricting their applications in broader engineering contexts. In contrast, deep learning-based surrogate models promise efficient, data-driven solutions. However, their effectiveness is often compromised by a reliance on deterministic frameworks, which fall short in accurately capturing the chaotic and stochastic nature of turbulence. The CoNFiLD model addresses these challenges by synergistically integrating conditional neural field encoding with latent diffusion processes, enabling the memory-efficient and robust probabilistic generation of spatiotemporal turbulence under varied conditions. Leveraging Bayesian conditional sampling, the model can seamlessly adapt to a diverse range of turbulence generation scenarios without the necessity for retraining, covering applications from zero-shot full-field flow reconstruction using sparse sensor measurements to super-resolution generation and spatiotemporal flow data restoration. Comprehensive numerical experiments across a variety of inhomogeneous, anisotropic turbulent flows with irregular geometries have been conducted to evaluate the model’s versatility and efficacy, showcasing its transformative potential in the domain of turbulence generation and the broader modeling of spatiotemporal dynamics.

[ Paper ]

Asynchronous Parallel Reinforcement Learning for Optimizing Propulsive Performance in Fin Ray Control

Published in Arxiv, 2024

[ Paper ]