Sciweavers

Explicit Manifold Representations for Value-Function Approximation in Reinforcement Learning