Model-Based Reinforcement Learning for Control under Time-Varying Dynamics

Model-Based Reinforcement Learning for Control under Time-Varying Dynamics explores An adaptive reinforcement learning algorithm that maintains control performance in systems with changing dynamics by intelligently managing historical data.. Commercial viability score: 4/10 in Model-Based RL.