Stiffness in Scientific Machine Learning: Cornell SCAN Seminar
Stiffness in Scientific Machine Learning
Chris Rackauckas (Mathematics, MIT)
Scientific machine learning is a burgeoning discipline for mixing machine learning into scientific simulation. Use cases of this field include automated discovery of physical equations and accelerating physical simulators. However, making the analyses of this field automated will require building a set of tools that handle stiff and ill-conditioned models without requiring user tuning. The purpose of this talk is to demonstrate how the methods and tools of scientific machine learning can be consolidated to give a single high performance and robust software stack. We will start by describing universal differential equations, a flexible mathematical object which is able to represent methodologies for equation discovery, 100-dimensional differential equation solvers, and discretizations of physics-informed neural networks. Then we will showcase how adjoint sensitivity analysis on the universal differential equation solving process gives rise to efficient and stiffly robust training methodologies for a large variety of scientific machine learning problems. With this understanding of differentiable programming we will describe how the Julia SciML Software Organization is utilizing this foundation to provide high performance tools for deploying battery powered airplanes, improving the energy efficiency of buildings, allow for navigation via the Earth’s magnetic field, and more.
http://www.cse.cornell.edu/scan/2020/12/07/rackauckas.html
https://diffeqflux.sciml.ai/dev/
Видео Stiffness in Scientific Machine Learning: Cornell SCAN Seminar канала Chris Rackauckas
Chris Rackauckas (Mathematics, MIT)
Scientific machine learning is a burgeoning discipline for mixing machine learning into scientific simulation. Use cases of this field include automated discovery of physical equations and accelerating physical simulators. However, making the analyses of this field automated will require building a set of tools that handle stiff and ill-conditioned models without requiring user tuning. The purpose of this talk is to demonstrate how the methods and tools of scientific machine learning can be consolidated to give a single high performance and robust software stack. We will start by describing universal differential equations, a flexible mathematical object which is able to represent methodologies for equation discovery, 100-dimensional differential equation solvers, and discretizations of physics-informed neural networks. Then we will showcase how adjoint sensitivity analysis on the universal differential equation solving process gives rise to efficient and stiffly robust training methodologies for a large variety of scientific machine learning problems. With this understanding of differentiable programming we will describe how the Julia SciML Software Organization is utilizing this foundation to provide high performance tools for deploying battery powered airplanes, improving the energy efficiency of buildings, allow for navigation via the Earth’s magnetic field, and more.
http://www.cse.cornell.edu/scan/2020/12/07/rackauckas.html
https://diffeqflux.sciml.ai/dev/
Видео Stiffness in Scientific Machine Learning: Cornell SCAN Seminar канала Chris Rackauckas
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
DifferentialEquations.jl: A performant and feature-rich ecosystem for solving differential equationsScientific Machine Learning (SciML) Helicopter Challenge ProblemWhat is (scientific) machine learning? An introduction through Julia's Lux.jlJulia and DifferentialEquations.jl : Chris Rackauckas, MITAccelerated Large-Eddy Simulation via Universal Partial Differential EquationsCOVID-19 Epidemic Mitigation via Scientific Machine Learning (SciML)Dense or Continuous Output for ODE SolversWhat is machine learning and how can it be connected to prior scientific knowledge (SciML)?Symbolics.jl - High performance symbolic numerics via multiple dispatch, Julia Computer AlgebraSimulation and Control of Biological Stochasticity - Chris Rackauckas PhD DefenseRunge-Kutta Methods for ODEsDifferential Equations in 2021Introduction to Scientific Machine Learning in Astroinformatics Part 2: NumericsPharmacometrics-Informed Deep Learning with DeepNLME - ISCB 2021 Invited SessionExtending Scientific Machine Learning (SciML) to Agent-Based Models (ICLR AI4ABM 2023)Model Discovery w/ Imposed Structures and Prior Knowledge Scientific Machine Learning | ML4ScienceAccelerating Quantitative Systems Pharmacology with Machine Learning - SMB 2021Automatic differentiation | New derivations of nonlinear solve and ODE adjointsStability-Optimized High Order Methods for Pathwise Stiffness in Stochastic Differential EquationsIntroduction to Scientific Machine Learning in Astroinformatics Part 1: Applications