The Dually Flat Geometry of Planning as Inference

arXiv:2609.04005v1 Announce Type: new Abstract: We present an alternative characterization of the occupancy measure of reinforcement learning, obtained by embedding the planning criterion into the dynamics through a resetting planning process. Its stationary measure, which we term visitation measure, is the object on which the information geometry of decision making is most naturally expressed. The achievable visitation measures form a dually flat statistical manifold whose two affine charts are the visitation probabilities and the log-policies, dual under the conditional entropy. This structure makes planning-as-inference generalize from linear rewards to nonlinear functionals of the visitation, each iterate solved by one natural-gradient step, and gives the temporal-difference error the interpretation of a marginal-utility estimate. We develop the geometry and its consequences for reinforcement learning and theoretical neuroscience.

Related

• Weight-Space Geometry of Offline Reasoning Training
• From Ticks to Flows: Dynamics of Neural Reinforcement Learning in Continuous Environments
• Exploiting Symbolic Heuristics for the Synthesis of Domain-Specific Temporal Planning Guidance using Reinforcement Learning
• Operator-on-F complements value-equivalence: a planning-time diagnostic for latent world models

Source: arXiv cs.AI | 2026-09-04