Abstract
Active Inference (AIF) is an emerging framework in the brain sciences which suggests that biological agents act to minimise a variational bound on model evidence. Control-as-Inference (CAI) is a framework within reinforcement learning which casts decision making as a variational inference problem. While these frameworks both consider action selection through the lens of variational inference, their relationship remains unclear. Here, we provide a formal comparison between them and demonstrate that the primary difference arises from how the notion of rewards, goals, or desires is incorporated into their generative models. We highlight how the encoding of value leads to subtle differences in the respective objective functionals and discuss how these distinctions lead to different exploratory behaviours.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have