DDPS|Generalizing Scientific Machine Learning and Differentiable Simulation Beyond Continuous models

Показать описание

Abstract: The combination of scientific models into deep learning structures, commonly referred to as scientific machine learning (SciML), has made great strides in the last few years in incorporating models such as ODEs and PDEs into deep learning through differentiable simulation. However, the vast space of scientific simulation also includes models like jump diffusions, agent-based models, and more. Is SciML constrained to the simple continuous cases or is there a way to generalize to more advanced model forms? This talk will dive into the mathematical aspects of generalizing differentiable simulation to discuss cases like chaotic simulations, differentiating stochastic simulations like particle filters and agent-based models, and solving inverse problems of Bayesian inverse problems (i.e. differentiation of Markov Chain Monte Carlo methods). We will then discuss the evolving numerical stability issues, implementation issues, and other interesting mathematical tidbits that are coming to light as these differentiable programming capabilities are being adopted.

IM release number: LLNL-VIDEO-856508

📲 Instagram: / livermore_lab
🐤 Twitter: / livermore_lab
🔔 Subscribe: / livermorelab

Inside Livermore Lab

Рекомендации по теме

Комментарии

Thanks for having me. It was a nice opportunity and I would be happy to follow up with anyone interested.

chrisrackauckasofficial

DDPS|Generalizing Scientific Machine Learning and Differentiable Simulation Beyond Continuous models

DDPS|Generalizing Scientific Machine Learning and Differentiable Simulation Beyond Continuous models

DDPS | Scientific Machine Learning through the Lens of Physics-Informed Neural Networks

DDPS | A flexible and generalizable XAI framework for scientific deep learning

DDPS | “Multi-fidelity linear regression for scientific machine learning from scarce data”

DDPS | The problem with deep learning for physics (and how to fix it) by Miles Cranmer

DDPS | Machine Learning and Multi-scale Modeling

DDPS | Model-constrained deep learning approaches for inference, control and UQ

DDPS | The passive symmetries of machine learning by Soledad Villar

DDPS | Machine Learning and Physics-based Simulations – Yin and Yang of Industrial Digit

DDPS | A mathematical understanding of modern Machine Learning: theory, algorithms and applications

DDPS | Reduced Order Modeling and Inverse Design of Flexible Structures by Machine Learning

Scientific Machine Learning: Physics-Informed Neural Networks with Craig Gin

DDPS | Generative Machine Learning Approaches for Data-Driven Modeling and Reductions

DDPS | Interpretable and Generalizable Machine Learning for Fluid Mechanics

DDPS | Physics-Guided Deep Learning for Dynamics Forecasting

DDPS | Non-intrusive reduced order models using physics informed neural networks

DDPS | Incorporating power system physics into deep learning via implicit layers

DDPS | Big Data Inverse Problems — Promoting Sparsity and Learning to Regularize by Mattias Chung

DDPS | Bayesian Optimization: Exploiting Machine Learning Models, Physics, & Throughput Experime...

DIRECT 2021 12 Scientific Machine Learning

DDPS |Scientific Uses of Automatic Differentiation by Michael Brenner

DDPS | Deep neural operators with reliable extrapolation for multiphysics & multiscale problems

DDPS | Neural Differentiable Physics

DDPS | Modeling and controlling turbulent flows through deep learning