12 papers
Expected free energy as an information constraint on the Bethe Lagrangian
Wouter M. Kouw
Active inference selects actions by minimising an expected free energy functional over predicted futures. However, adding an expectation over yet-unobserved outcomes means the free…
On the abstract elementary class of acts with pure embeddings
Jonathan Feigert, Daniel Herden, Marcos Mazari-Armida
We study the abstract elementary class of acts with pure embeddings. In particular, we show that stability and superstability in this class can be characterized in terms of the mon…
Expected Free Energy-based Planning as Variational Inference
Wouter W. L. Nuijten, Thijs van de Laar, Bert de Vries
Planning under uncertainty requires agents to balance goal achievement with information gathering. Active inference addresses this through the Expected Free Energy (EFE), a cost fu…
Composing Non-Conjugate Factor Graphs with Closed-Form Variational Inference
Mykola Lukashchuk, Kyrylo Yemets, Wouter M. Kouw +4
Stacking probabilistic building blocks into deeper architectures typically breaks closed-form inference. We show that closed-form inference can be preserved. We identify five facto…
Curvature-aware Expected Free Energy as an Acquisition Function for Bayesian Optimization
Ajith Anil Meera, Wouter Kouw
We propose an Expected Free Energy-based acquisition function for Bayesian optimization to solve the joint learning and optimization problem, i.e., optimize and learn the underlyin…
Gaussian Variational Inference with Non-Gaussian Factors for State Estimation: A UWB Localization Case Study
Andrew Stirling, Mykola Lukashchuk, Dmitry Bagaev +2
This letter extends the exactly sparse Gaussian variational inference (ESGVI) algorithm for state estimation in two complementary directions. First, ESGVI is generalized to operate…