GOVERNED BY PHYSICS, DRIVEN BY DATA: ORCHESTRATING THE SMART PROTOTYPE LIFECYCLE WITH AI AGENTS
Wednesday 20 May 2026 14:20
How do you build a "smart" prototype when data is scarce and failure is not an option? Standard AI often fails because it ignores the laws of nature. This 5-minute session introduces Scientific Machine Learning (SciML) – a hybrid approach that embeds first-principle physics directly into neural networks.
We will rapidly cover how to Reduce Data Dependency (use physics to "fill the gaps," allowing for high-fidelity models with fewer sensors): Ensure Physical Integrity (guarantee that conservation of energy and mass are respected by design, not by chance); Automate with AI Agents (leverage autonomous agents to bridge the gap between physical measurements and digital twins). Discover how to transform your testing from a data-collection exercise into a self-evolving, physically intelligent system.
We will rapidly cover how to Reduce Data Dependency (use physics to "fill the gaps," allowing for high-fidelity models with fewer sensors): Ensure Physical Integrity (guarantee that conservation of energy and mass are respected by design, not by chance); Automate with AI Agents (leverage autonomous agents to bridge the gap between physical measurements and digital twins). Discover how to transform your testing from a data-collection exercise into a self-evolving, physically intelligent system.
