Exploring and learning structure : active inference approach in navigational agents
(2025)
ACTIVE INFERENCE, IWAI 2024.
In Communications in computer and information science
2193.
p.105-118
- Author
- Daria de Tinguy (UGent) , Tim Verbelen (UGent) and Bart Dhoedt (UGent)
- Organization
- Project
- Abstract
- Drawing inspiration from animal navigation strategies, we introduce a novel computational model for navigation and mapping, rooted in biologically inspired principles. Animals exhibit remarkable navigation abilities by efficiently using memory, imagination, and strategic decision-making to navigate complex and aliased environments. Building on these insights, we integrate traditional cognitive mapping approaches with an Active Inference Framework (AIF) to learn an environment structure in a few steps. Through the incorporation of topological mapping for long-term memory and AIF for navigation planning and structure learning, our model can dynamically apprehend environmental structures and expand its internal map with predicted beliefs during exploration. Comparative experiments with the Clone-Structured Graph (CSCG) model highlight our model’s ability to rapidly learn environmental structures in a single episode, with minimal navigation overlap. This is achieved without prior knowledge of the dimensions of the environment or the type of observations, showcasing its robustness and effectiveness in navigating ambiguous environments.
- Keywords
- exploration, active inference, topological graph, structure learning, COGNITIVE MAP
Downloads
-
(...).pdf
- full text (Accepted manuscript)
- |
- UGent only
- |
- |
- 3.86 MB
-
(...).pdf
- full text (Accepted manuscript)
- |
- UGent only (changes to open access on 2027-01-01)
- |
- |
- 1.11 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01JGXGANZBJ1HTRDWC52QFR60T
- MLA
- de Tinguy, Daria, et al. “Exploring and Learning Structure : Active Inference Approach in Navigational Agents.” ACTIVE INFERENCE, IWAI 2024, edited by Christopher L. Buckley et al., vol. 2193, Springer, 2025, pp. 105–18, doi:10.1007/978-3-031-77138-5_7.
- APA
- de Tinguy, D., Verbelen, T., & Dhoedt, B. (2025). Exploring and learning structure : active inference approach in navigational agents. In C. L. Buckley, D. Cialfi, P. Lanillos, R. J. Pitliya, N. Sajid, H. Shimazaki, … M. Wisse (Eds.), ACTIVE INFERENCE, IWAI 2024 (Vol. 2193, pp. 105–118). https://doi.org/10.1007/978-3-031-77138-5_7
- Chicago author-date
- Tinguy, Daria de, Tim Verbelen, and Bart Dhoedt. 2025. “Exploring and Learning Structure : Active Inference Approach in Navigational Agents.” In ACTIVE INFERENCE, IWAI 2024, edited by Christopher L. Buckley, Daniela Cialfi, Pablo Lanillos, Riddhi J. Pitliya, Noor Sajid, Hideaki Shimazaki, Tim Verbelen, and Martijn Wisse, 2193:105–18. Cham: Springer. https://doi.org/10.1007/978-3-031-77138-5_7.
- Chicago author-date (all authors)
- de Tinguy, Daria, Tim Verbelen, and Bart Dhoedt. 2025. “Exploring and Learning Structure : Active Inference Approach in Navigational Agents.” In ACTIVE INFERENCE, IWAI 2024, ed by. Christopher L. Buckley, Daniela Cialfi, Pablo Lanillos, Riddhi J. Pitliya, Noor Sajid, Hideaki Shimazaki, Tim Verbelen, and Martijn Wisse, 2193:105–118. Cham: Springer. doi:10.1007/978-3-031-77138-5_7.
- Vancouver
- 1.de Tinguy D, Verbelen T, Dhoedt B. Exploring and learning structure : active inference approach in navigational agents. In: Buckley CL, Cialfi D, Lanillos P, Pitliya RJ, Sajid N, Shimazaki H, et al., editors. ACTIVE INFERENCE, IWAI 2024. Cham: Springer; 2025. p. 105–18.
- IEEE
- [1]D. de Tinguy, T. Verbelen, and B. Dhoedt, “Exploring and learning structure : active inference approach in navigational agents,” in ACTIVE INFERENCE, IWAI 2024, Oxford, UK, 2025, vol. 2193, pp. 105–118.
@inproceedings{01JGXGANZBJ1HTRDWC52QFR60T,
abstract = {{Drawing inspiration from animal navigation strategies, we introduce a novel computational model for navigation and mapping, rooted in biologically inspired principles. Animals exhibit remarkable navigation abilities by efficiently using memory, imagination, and strategic decision-making to navigate complex and aliased environments. Building on these insights, we integrate traditional cognitive mapping approaches with an Active Inference Framework (AIF) to learn an environment structure in a few steps. Through the incorporation of topological mapping for long-term memory and AIF for navigation planning and structure learning, our model can dynamically apprehend environmental structures and expand its internal map with predicted beliefs during exploration. Comparative experiments with the Clone-Structured Graph (CSCG) model highlight our model’s ability to rapidly learn environmental structures in a single episode, with minimal navigation overlap. This is achieved without prior knowledge of the dimensions of the environment or the type of observations, showcasing its robustness and effectiveness in navigating ambiguous environments.}},
author = {{de Tinguy, Daria and Verbelen, Tim and Dhoedt, Bart}},
booktitle = {{ACTIVE INFERENCE, IWAI 2024}},
editor = {{Buckley, Christopher L. and Cialfi, Daniela and Lanillos, Pablo and Pitliya, Riddhi J. and Sajid, Noor and Shimazaki, Hideaki and Verbelen, Tim and Wisse, Martijn}},
isbn = {{9783031771378}},
issn = {{1865-0929}},
keywords = {{exploration,active inference,topological graph,structure learning,COGNITIVE MAP}},
language = {{eng}},
location = {{Oxford, UK}},
pages = {{105--118}},
publisher = {{Springer}},
title = {{Exploring and learning structure : active inference approach in navigational agents}},
url = {{http://doi.org/10.1007/978-3-031-77138-5_7}},
volume = {{2193}},
year = {{2025}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: