VSA RL algorithm for Gridworld. Uses Double Q Learning and a target model to enhance convergence stability. Uses relational observations space. One randomly placed avoidance and one randomly placed goal each episode.
HowardGoldowsky/Vector_Symbolic_RL_Gridworld
VSA RL algorithm for Gridworld. Uses Double Q Learning and a target model to enhance convergence stability. Uses relational observations space. One randomly placed avoidance and one randomly placed goal each episode.
MATLAB