2025-07-22 –, Main Room 2
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!
This workshop is more designed as an interactive challenge. We will use the Julia-based SpeedyWeather.jl / RainMaker.jl as a model and participates will participate in a challenge to find parameter solutions that cause the most rain in a specific location. Premade scripts that show how to run the weather model will be provided. The goal will be to find parameters of the model which maximize the rainfall.
Solutions which start by building an ML surrogate model will be shown to the participants as a starting point. A live interactive leaderboard automatically updated through CI (Documenter.jl builds) will be used in order track the progress of different participants. Suggested techniques, using tools like LIBSVM.jl, XGBoost.jl, etc. will be discussed and participants will try various ML and SciML tools to achieve the goals. At the end, the participants will share their solutions and a "winner" to the challenge will be crowned.
The goal of course is not to win, but the friends you make along the way. By the end, participants will have had fun while learning new ML tools within Julia, gained experience mixing climate/weather modeling tools with ML, and have discussions about what new tooling could further enhance the workflows.
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.
