NCHC AI Powered Physics Bootcamp, 2026

Overview / 活動說明

Together with NVIDIA and OpenACC organization will host an online bootcamp on March 10, 2026.

The AI-Powered Physics Bootcamp helps participants revolutionize their approach to computational physics problems by combining traditional physics-based modeling with modern artificial intelligence (AI) to tackle real-world scientific computing challenges orders of magnitude faster than traditional simulation methods.

Attendees are introduced to NVIDIAR PhysicsNeMo, an open-source Python framework for building, training, and fine-tuning physics AI models, and guided through progressive challenges to solve complex scientific problems by combining physics-based partial differential equations (PDEs) with AI.

Hands-on practical skills are gained by participants to speed up their research and engineering workflows, whether they are optimizing designs, forecasting complex phenomena, or solving multi-physics problems. Afterwards, the attendees will have the confidence to apply physics-informed AI to their domain, comprehend when to use physics constraints versus data-driven approaches, and how to visualize and validate their results.

By attending this bootcamp, participants will:
· Learn the fundamental differences between physics-driven and data-driven approaches to neural networks;
· Train more robust and generalizable AI models by leveraging physical laws as constraints;
· Understand how to parameterize and solve inverse problems;
· Gain practical experience visualizing simulation results using ParaView; and
· Learn advanced topics such as DoMINO and Earth-2.

Timeline / 活動期程
· Event Date / 本次活動日期:  March 10–11, 2026
· Registration Deadline / 報名截止日期:  February 17, 2026

Registraion OpenACC Website / 活動報名網站:
https://www.openhackathons.org/s/siteevent/a0CUP00003pPKYF2A4/se000470

Notes / 注意事項
· The event is limited to those living in Taiwan, Penghu, Kinmen, and Matsu and will be entirely in Chinese.
本次活動僅限居住在臺灣、澎湖、金門、馬祖地區的人參加，並完全以中文進行。
· To receive notifications for this event, please ensure that you leave a valid email address when applying.
本次活動的所有通知訊息都將透過 email 發送，請在報名時務必留下可以聯絡到的 email 信箱。

Agenda / 活動規劃 (如有變動以活動官網說明為準)

March 10 (Tuesday, Online event through MS Teams)

March 11 (Wednesday, Physical event at NCHC)

·       09:00 AM – 11:00 AM Welcome and Getting Started with PhysicsNeMo o   Overview of Deep Learning fundamentals o   Overview of PhysicsNeMo framework and workflow o   Physics-Informed vs Data-driven approaches ·       11:00 AM – 11:30 PM Access to machines o   Access to cluster and setup the bootcamp environment ·       11:30 AM – 12:00 PM Lab 1. Introduction to Physics-Informed Neural Networks (PINNs) o   Neural Network Solver Methodology o   Parameterized Problems and Inverse Problems

·       09:30 AM – 10:00 AM Lab 2. Solving ODEs with PhysicsNeMo o   Simulating Projectile Motion ODE o   Introduction to ParaView visualization ·       10:00 AM – 10:30 AM Lab 3. From ODEs to PDEs - Diffusion Problems o   Steady State 1D Diffusion in a Composite Bar o   Parameterized 1D Diffusion with interface conditions ·       10:30 AM – 11:00 AM Lab 4. Advanced PDE Systems - Navier-Stokes o   Forecasting weather at sea level using Navier-Stokes equations ·       11:00 PM – 12:00 PM Challenge 1. Advanced Wave Dynamics o   Level 1: Basic 2D wave equation o   Level 2: Variable wave speed in heterogeneous media o   Level 3: Multiple sources and complex boundary conditions ·       12:00 PM – 13:00 PM     Lunch Break ·       13:00 PM – 14:00 PM Challenge 2. Fluid-Structure Interaction (Chip Cooling) o   Level 1: Single chip with fixed geometry o   Level 2: Multiple chips with flow optimization o   Level 3: Coupled flow-thermal problem with design constraints ·       14:00 PM – 15:00 PM Challenge 3. Multi-Physics Climate Modeling o   Level 1: Simple atmosphere model with heat diffusion o   Level 2: Atmosphere-ocean coupling ·       15:00 PM – 16:00 PM Challenge 4. Advanced Neural Operators o   Level 1: Basic FNO (Fourier Neural Operator) implementation o   Level 2: AFNO and PINO comparison o   Level 3: Custom architecture design and optimization ·       16:00 PM      Wrap-up, Q&A