Year: 2025
Type: preprint
Abstract: We present the design of an autoregressive active inference agent in the form of message passing on a factor graph. Expected free energy is derived and distributed across a planning graph. The proposed agent is validated on a robot navigation task, demonstrating exploration and exploitation in a continuous-valued observation space with bounded continuous-valued actions. Compared to a classical optimal controller, the agent modulates action based on predictive uncertainty, arriving later but with a better model of the robot's dynamics.
Source: arXiv (Cornell University)
Authors: Kouw, Wouter M., Nisslbeck, Tim N., Nuijten, Wouter L. N.
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7084626052
Cites:
Cited by:
Related to: