Message passing-based inference in an autoregressive active inference agent
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912831035670528 |
|---|---|
| author | Kouw, Wouter M. Nisslbeck, Tim N. Nuijten, Wouter L. N. |
| author_facet | Kouw, Wouter M. Nisslbeck, Tim N. Nuijten, Wouter L. N. |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25482 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Message passing-based inference in an autoregressive active inference agent Kouw, Wouter M. Nisslbeck, Tim N. Nuijten, Wouter L. N. Artificial Intelligence Machine Learning Robotics Systems and Control 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. |
| title | Message passing-based inference in an autoregressive active inference agent |
| topic | Artificial Intelligence Machine Learning Robotics Systems and Control |
| url | https://arxiv.org/abs/2509.25482 |