Axiom, developed by Verse AI, represents a breakthrough in machine learning by incorporating human-like cognitive modeling through a foundational understanding of physical interactions in game worlds. Unlike traditional deep reinforcement learning which requires extensive experimentation, Axiom utilizes active inference to update its model based on observed game behaviors, dramatically enhancing efficiency. This innovative approach is inspired by the free energy principle, a theoretical framework from neuroscience posited by Karl Friston that aims to resemble actual brain cognition. Axiom's capacity to master various simplified video games with minimal resources sets it apart in AI development.
The new system, called Axiom, offers an alternative to the artificial neural networks that are dominant in modern AI.
Friston told me over video from his home in London that the approach may be especially important for building AI agents.
Collection
[
|
...
]