Chris Rackauckas
Dr. Chris Rackauckas is the VP of Modeling and Simulation at JuliaHub, the Director of Scientific Research at Pumas-AI, Co-PI of the Julia Lab at MIT, and the lead developer of the SciML Open Source Software Organization. For his work in mechanistic machine learning, his work is credited for the 15,000x acceleration of NASA Launch Services simulations and recently demonstrated a 60x-570x acceleration over Modelica tools in HVAC simulation, earning Chris the US Air Force Artificial Intelligence Accelerator Scientific Excellence Award. See more at https://chrisrackauckas.com/. He is the lead developer of the Pumas project and has received a top presentation award at every ACoP in the last 3 years for improving methods for uncertainty quantification, automated GPU acceleration of nonlinear mixed effects modeling (NLME), and machine learning assisted construction of NLME models with DeepNLME. For these achievements, Chris received the Emerging Scientist award from ISoP.
Sessions
DifferentialEquations.jl is the main package in Julia for solving differential equations. It has all sorts of things, from solving ordinary differential equations to stochastic differential equations, differential-algebraic equations, and more. You can switch to preconditioned GMRES linear solvers, exponential integrators, integrate with automatic differentiation, tweak the nonlinear solvers, and add customized stepping/logging routines. Most people only ever scratch the surface: let's dive in.
Building surrogates of fluid dynamics models is a common way to accelerate the analyses. In this workshop we will go hands-on with ML tooling to improve the ability to analyze a weather model. A live challenge to find the parameters that maximize rainfall in a given model will drive the discussion. Participants will interact with the model and submit solutions to a leaderboard to crown a winner. No prior ML or weather modeling experience required!
The Julia SciML solvers in DifferentialEquations.jl are already pretty optimized for stiff ODEs and DAEs, so where does the next order of magnitude in performance come from? In this talk we will describe how symbolic-numeric compiler tricks are being integrated in to the solver architecture in order to achieve performance that is beyond anything possible with purely numerical systems.
Solving nonlinear least squares problems is very common across many aspects of science, from the implementation of curve fitting in data analysis to complex nonlinear optimizations. In this talk we will talk about the latest advancements in solving nonlinear least squares problems in Julia. This includes a discussion of the newest methods and packages, along with the remaining challenges around discoverability and documentation.
Comments, questions, or concerns about the future direction of the SciML tools? Come here to discuss what we want to see in the near future.