Oct 25, 2021

Embodied intelligence via learning and evolution

Posted by in categories: evolution, physics

The authors present a high-resolution palaeomagnetic record for a Late Cretaceous limestone in Italy. They claim that their record robustly shows a ~12° true polar wander oscillation between 86 and 78 Ma, with the greatest excursion at 84–82 Ma.

The authors propose a new framework, deep evolutionary reinforcement learning, evolves agents with diverse morphologies to learn hard locomotion and manipulation tasks in complex environments, and reveals insights into relations between environmental physics, embodied intelligence, and the evolution of rapid learning.

Leave a reply