AI learns the cosmos, but bias may hide new physics
TL;DR
Researchers trained an AI on standard ΛCDM cosmology simulations and tested its ability to probe beyond the standard model. Transfer learning reduced the number of simulations needed, but the AI developed negative transfer biases, mistaking known patterns for new effects and risking missed clues about new physics. The team emphasizes careful interpretation and plans to test the approach on more realistic survey data to determine when AI can reliably accelerate cosmological discovery.
Topics:science#artificial-intelligence#cosmology#machine-learning#negative-transfer#science#transfer-learning
- AI Learned How the Universe Works—and That Created an Unexpected Problem for Physicists Gizmodo
- To discover new physics, AI may need to 'unlearn' the old one Phys.org
- Artificial intelligence requires unlearning to discover new physics laws Open Access Government
- Transfer Learning Slashes Cosmology AI Costs: Neutrino Mass Degeneracy Triggers Negative Transfer Tech Times
- AI could uncover new physics faster but there’s a surprising catch ScienceDaily
Want the full story? Read the original reporting
Read on Gizmodo