ScienceLanguage: English
Physics Model Reveals Limits of AI Energy Efficiency
Key Takeaways
- Discovered a fundamental trade-off between AI prediction and energy efficiency.
- Led by Hans Briegel at the University of Innsbruck.
- Published in the scientific journal Physical Review X.
- Applies theoretical physics to understand computational thermodynamic limits.
Researchers at the University of Innsbruck have demonstrated a core trade-off in intelligent systems, showing that higher prediction accuracy demands greater energy expenditure. Using theoretical physics, the team analyzed the physical limits of information processing in artificial intelligence. Their findings shed light on the thermodynamic costs of running complex algorithms and robotic systems in the real world.
Recommended for you
Tools and services we trust to boost productivity and content workflows.
Browse picks