Deep Deterministic Policy Gradient

The algorithm for Deep Deterministic Policy Gradient (DDPG) is given as follows:

1. Initialize the main critic network parameter and main actor network parameter
2. Initialize the target critic network parameter by just copying the main critic network parameter
3. Initialize the target actor network parameter by just copying the main actor network parameter .
4. Initialize the replay buffer
5. For N number of episodes, repeat steps 6 to 7
6. Initialize an Ornstein-Uhlenbeck random process for action space exploration
7. For each step in the episode, that is, for t = 0, . . ., T – 1:
   1. Select action a based on the policy and exploration noise, that is,
   2. Perform the selected action a, move to the next state and get the reward r, and store this transition information in the replay buffer
   3. Randomly sample a minibatch of K transitions from the replay buffer
   4. Compute the target...