This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
Active Inference with Dynamic Planning and Information Gain in Continuous Space by Inferring Low-Dimensional Latent States
1
Cognitive Neurorobotics Research Unit, Okinawa Institute of Science and Technology, Okinawa 904-0495, Japan
2
Department of Electronics and Electrical Engineering, Keio University, Kanagawa 223-8522, Japan
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Entropy 2025, 27(8), 846; https://doi.org/10.3390/e27080846 (registering DOI)
Submission received: 21 June 2025
/
Revised: 1 August 2025
/
Accepted: 7 August 2025
/
Published: 9 August 2025
Abstract
Active inference offers a unified framework in which agents can exhibit both goal-directed and epistemic behaviors. However, implementing policy search in high-dimensional continuous action spaces presents challenges in terms of scalability and stability. Our previously proposed model, T-GLean, addressed this issue by enabling efficient goal-directed planning through low-dimensional latent space search, further reduced by conditioning on prior habituated behavior. However, the lack of an epistemic term in minimizing expected free energy limited the agent’s ability to engage in information-seeking behavior that can be critical for attaining preferred outcomes. In this study, we present EFE-GLean, an extended version of T-GLean that overcomes this limitation by integrating epistemic value into the planning process. EFE-GLean generates goal-directed policies by inferring low-dimensional future posterior trajectories while maximizing expected information gain. Simulation experiments using an extended T-maze task—implemented in both discrete and continuous domains—demonstrate that the agent can successfully achieve its goals by exploiting hidden environmental information. Furthermore, we show that the agent is capable of adapting to abrupt environmental changes by dynamically revising plans through simultaneous minimization of past variational free energy and future expected free energy. Finally, analytical evaluations detail the underlying mechanisms and computational properties of the model.
Share and Cite
MDPI and ACS Style
Matsumoto, T.; Fujii, K.; Murata, S.; Tani, J.
Active Inference with Dynamic Planning and Information Gain in Continuous Space by Inferring Low-Dimensional Latent States. Entropy 2025, 27, 846.
https://doi.org/10.3390/e27080846
AMA Style
Matsumoto T, Fujii K, Murata S, Tani J.
Active Inference with Dynamic Planning and Information Gain in Continuous Space by Inferring Low-Dimensional Latent States. Entropy. 2025; 27(8):846.
https://doi.org/10.3390/e27080846
Chicago/Turabian Style
Matsumoto, Takazumi, Kentaro Fujii, Shingo Murata, and Jun Tani.
2025. "Active Inference with Dynamic Planning and Information Gain in Continuous Space by Inferring Low-Dimensional Latent States" Entropy 27, no. 8: 846.
https://doi.org/10.3390/e27080846
APA Style
Matsumoto, T., Fujii, K., Murata, S., & Tani, J.
(2025). Active Inference with Dynamic Planning and Information Gain in Continuous Space by Inferring Low-Dimensional Latent States. Entropy, 27(8), 846.
https://doi.org/10.3390/e27080846
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article Access Statistics
For more information on the journal statistics, click
here.
Multiple requests from the same IP address are counted as one view.