AI in the physical world: from observation to discovery
summary
Next‑generation gamma‑ray array spans 50 telescopes in two hemispheres, producing 10 ns camera exposures and about 6 PB/year of raw data; engineers use deep learning to reconstruct air‑shower events and LLMs/agents for code, documentation and pipeline design. A reproduction experiment replaces a human pipeline with an agent that iterates models (random forests baseline then agent improved classifier), showing stronger automated tuning but persistent simulation‑to‑real mismatches and brittle operational integration.
AI-written summary · about
chapters (6)
- 0:00Introduction and background
- 2:13AI in scientific process
- 10:42LLMs in physics research
- 14:40AI replacing scientists
- 16:49Future of AI in science
- 19:01Q&A
related talks
38mAI in action: How to use practically in iGaming beyond the hype?
16mAI Powering Science Breakthroughs
46mPath of Data & AI Challenges
13mBuilding High-Performance AI Literature Processing
10mRadiological AGI: How Close Are We?
14mAI for Science
47mBeyond Agents: What AI Strategy Really Needs in 2025
38mAI in Kubernetes: How to Get Started?
