Joint-Space Empowerment as a Theory of Dexterous Motor Coordination

Searching for effective policies is notoriously challenging in overactuated tendon-driven systems, where each joint is actuated by many muscles or motorized cables. Although this redundancy complicates naive policy search, it also implies that effective control can be captured by a low-dimensional action manifold. To identify such a manifold, we introduce Joint-Space Empowerment (JoSE), a novel information-theoretic objective that quantifies how much control an agent has over its mechanical degrees of freedom. We frame manifold discovery as an optimal precoding problem-where a state-dependent precoder maps low-dimensional latent actions to high-dimensional actions-and derive its closedform solution under learned control-affine Gaussian dynamics. Across both a musculoskeletal hand model and a tendon-driven robotic hand, we show that policies trained on this manifold achieve significantly enhanced dexterity, sample efficiency, and improved generalization. More broadly, these results present optimal precoding as a general information-theoretic paradigm for coordinating high-dimensional actuators to control low-dimensional features. Project page: https://joint-space-empowerment.github.io.