Decomposing Control Lyapunov Functions for Efficient Reinforcement Learning

Using Decomposed Control Lyapunov Functions can reduce sample complexity and improve RL performance.