Active Inference or Control as Inference? A Unifying View
This work provides a unifying theoretical view for researchers in computational neuroscience and control theory, but it is incremental as it builds on existing frameworks without demonstrating new practical algorithms.
The paper tackles the challenge of making active inference (AI) practical for sensorimotor control by framing it through control as inference (CaI), showing that AI can be viewed as partially-observed CaI with cost functions defined in observation states.
Active inference (AI) is a persuasive theoretical framework from computational neuroscience that seeks to describe action and perception as inference-based computation. However, this framework has yet to provide practical sensorimotor control algorithms that are competitive with alternative approaches. In this work, we frame active inference through the lens of control as inference (CaI), a body of work that presents trajectory optimization as inference. From the wider view of `probabilistic numerics', CaI offers principled, numerically robust optimal control solvers that provide uncertainty quantification, and can scale to nonlinear problems with approximate inference. We show that AI may be framed as partially-observed CaI when the cost function is defined specifically in the observation states.