Explore the frontiers of scientific machine learning (SciML) and differentiable simulation in this 58-minute lecture from the Inside Livermore Lab series. Delve into the mathematical aspects of generalizing differentiable simulation beyond continuous models, examining cases such as chaotic simulations, stochastic simulations like particle filters and agent-based models, and solving inverse problems of Bayesian inverse problems. Discover the evolving numerical stability issues, implementation challenges, and intriguing mathematical nuances emerging as differentiable programming capabilities gain wider adoption. Learn from Dr. Chris Rackauckas, a leading expert in mechanistic machine learning, as he shares insights from his work on accelerating NASA Launch Services simulations and HVAC simulation. Gain valuable knowledge about the expanding scope of SciML and its potential applications in various scientific domains.
Generalizing Scientific Machine Learning and Differentiable Simulation Beyond Continuous Models
Inside Livermore Lab via YouTube
Overview
Syllabus
DDPS|Generalizing Scientific Machine Learning and Differentiable Simulation Beyond Continuous models
Taught by
Inside Livermore Lab
Related Courses
-
Generalizing Scientific Machine Learning and Differentiable Simulation Beyond Continuous Models
-
Scientific Computing and Machine Learning for Industrial Modeling - An Introduction to SciML
-
Neural Differentiable Physics: Unifying Numerical PDEs and Deep Learning for Data-Augmented Computational Physics
-
Infinite Dimensional Optimization for Scientific Machine Learning
-
Scientific Machine Learning through the Lens of Physics-Informed Neural Networks
-
Machine Learning for Computational Fluid Dynamics
