Explore a framework for predictive modeling of unknown systems using measurement data in this 1 hour 15 minute lecture by Dongbin Xiu. Learn how deep neural networks are employed to discover and approximate unknown evolution operators, enabling system analysis and prediction. Discover the applications of flow map learning (FML) in modeling dynamical systems, systems with missing variables and hidden parameters, and partial differential equations (PDEs). Gain insights from Xiu's extensive research background in uncertainty quantification, stochastic computing, and machine learning as he presents this innovative approach to data-driven modeling.
Overview
Syllabus
DDPS | Data-driven modeling of unknowns systems with deep neural networks by Dongbin Xiu
Taught by
Inside Livermore Lab
Related Courses
-
Probabilistic Deep Learning with TensorFlow 2
-
Derivative-Informed Neural Operators for Solving PDE-Governed Optimization Problems
-
DeepParticle: Learning Invariant Measure by Deep Neural Network Minimizing Wasserstein Distance
-
Deep Learning and Computations of PDEs by Siddhartha Mishra
-
Leveraging Latent Representations for Predictive Physics-Based Modeling
-
Neural Differentiable Physics: Unifying Numerical PDEs and Deep Learning for Data-Augmented Computational Physics
