img-blur-shadow-blog-2

Our lab's research solves the core and universal problems at the intersection of AI (deep learning) and scientific disciplines, including:

(1) Generative AI for scientific simulation, control, and design: Develop generative AI methods based on diffusion models, flow matching, and next-generation generative models for the simulation, control, and design of large-scale scientific systems. Applications include critical fields such as fluid dynamics, energy systems, mechanical engineering, and life sciences.

(2) AI Agents for scientific discovery: Create machine learning approaches integrating AI agents, representation learning, and foundation models to build general-purpose AI scientists. These systems will enable intelligent automation of scientific experiments and generate new scientific discoveries, with applications in life sciences and physics.

(3) Representation learning with graph neural networks, generative models, and information theory, motivated by scientific applications.

img-blur-shadow-blog-2
Publications

CL-DiffPhyCon: Closed-loop Diffusion Control of Complex Physical Systems

We propose a diffusion method with an asynchronous denoising schedule for physical systems control tasks. It achieves closed-loop control with significant speedup of sampling efficiency.Read More

Invited Talk


[2025/04] 生成模型用于复杂系统的设计与控制以及test-time scaling,at 北大前沿探索者沙龙

[2024/12] AI+科学和工程:从2025到2035,at 浦江AI学术年会

[2024/10] 基于扩散生成模型的复杂系统设计和控制, at CNCC 2024


[2023/05] Graph Neural Networks for Large-scale Scientific Simulations, at 首届人工智能科学计算研讨会

[2023/03] Steps toward an AI scientist: neuro-symbolic models for concept generalization and theory learning, at AAAI 2023 Symposium of Computational Approaches to Scientific Discovery

slides

[2025/04] 生成模型用于复杂系统的设计与控制以及test-time scaling,at 北大前沿探索者沙龙

[2024/12] AI+科学和工程:从2025到2035,at 浦江AI学术年会

[2024/10] 基于扩散生成模型的复杂系统设计和控制, at CNCC 2024

[2024/10] Learning adaptive and compositional networks for multi-resolution simulation and inverse design, at IMS-NTU joint workshop on Applied Geometry for Data Sciences, Part I

[2024/07] Designing and controlling complex systems via diffusion generative models, at the 10th Shanghai International Symposium on Nonlinear Sciences and Applications

[2024/04] Envisioning AI + Science in 2030, at 2050 Yunqi Forum

[2024/03] Uncertainty Quantification for Forward and Inverse Problems of PDEs via Latent Global Evolution, at SIAM Conference on Uncertainty Quantification (UQ24)

[2024/02] Compositional generative inverse design, at AI4Science Talks

slide

[2024/01] Machine learning for accelerating scientific simulation, design and control, at Annual Meeting of Peking University Center for Quantitative Biology

[2023/11] GNN for scientific simulations: towards adaptive multi-resolution simulators and a foundation neural operator, at LoG Shanghai Meetup

[2023/08] Graph neural network for accelerating scientific simulation, at 2023 科学智能峰会

[2023/08] Phase transitions in the universal tradeoff between accuracy and simplicity in machine learning, at HK Satellite of StatPhys28

[2023/07] 走向AI物理学家,做科学发现和仿真, at 果壳未来光锥

[2023/06] Learning structured representations for accelerating simulation and design, at School of Intelligence Science and Technology, Peking University

[2023/05] Steps toward an AI scientist: neuro-symbolic models for concept generalization and theory learning, at Brown University Autonomous Empirical Research group.

[2023/04] Learning Controllable Adaptive Simulation for Multi-resolution Physics, at Stanford Data for Sustainability Conference 2023

slides

[2023/04] AI for scientific design: Surrogate model + backpropagation method, at Swarma

slides

[2023/03] Learning Controllable Adaptive Simulation for Multi-resolution Physics, at TechBeat

slides video

[2023/02] ZeroC: A Neuro-Symbolic Model for Zero-shot Concept Recognition and Acquisition at Inference Time, at AI Time

slides

[2023/02] Graph Neural Networks for Large-scale Simulations, at Stanford HAI Climate-Centred Student Affinity Group

slides

[2022/12] Steps toward an AI scientist: neuro-symbolic models for zero-shot learning of concepts and theories, at BIGAI

[2022/09] Learning to Accelerate Large-Scale Physical Simulations in Fluid and Plasma Physics, at SIAM 2022 Conference on Mathematics of Data Science

[2022/06] Learning to accelerate simulation and inverse optimization of PDEs via latent global evolution, Stanford CS ML lunch

[2021/12] Machine learning of physics theories, at Alan Turing Institute, Machine Learning and Dynamical Systems Seminar

video

[2021/10] Graph Information Bottleneck, at Beijing Academy of Artificial Intelligence (BAAI)

[2021/06] Machine learning of physics theories and its universal tradeoff between accuracy and simplicity, at Workshop on Artificial Scientific Discovery 2021

slides video

[2021/04] Phase transitions on the tradeoff between prediction and compression in machine learning, Stanford CS ML lunch

[2021/04] Machine Learning of Physics Theories, at Workshop on Artificial Scientific Discovery 2021, Summer school hosted by Max Planck Institute for the Science of Light

slides video

[2021/02] Machine Learning of Physics Theories, at Seminar Series of SJTU Institute of Science

video

[2020/10] Machine learning of physics theories and its universal tradeoff between accuracy and simplicity, at Los Alamos National Lab

[2020/02] Phase transitions for the information bottleneck, UIUC

News