Rete del Critic

I metodi Actor-Critic richiedono due reti neurali molto diverse.

L’architettura della rete dell’actor è identica a quella della policy network che hai usato in REINFORCE, quindi puoi riutilizzare la classe PolicyNetwork.

Tuttavia, la rete del critic è qualcosa che finora non hai implementato. Il critic mira ad approssimare la funzione valore di stato \(V(s_t)\), invece della funzione valore azione \(Q(s_t, a_t)\) approssimata dalle Q-Network.

Ora implementerai il modulo di rete Critic che userai in A2C.

Questo esercizio fa parte del corso

Deep Reinforcement Learning in Python

Visualizza il corso

Istruzioni dell'esercizio

Inserisci la dimensione desiderata per il secondo layer fully connected in modo che produca un solo valore di stato.
Ottieni il valore restituito dal forward pass attraverso la rete del critic.

Esercizio pratico interattivo

Prova a risolvere questo esercizio completando il codice di esempio.

class Critic(nn.Module):
    def __init__(self, state_size):
        super(Critic, self).__init__()
        self.fc1 = nn.Linear(state_size, 64)
        # Fill in the desired dimensions
        self.fc2 = nn.Linear(____)

    def forward(self, state):
        x = torch.relu(self.fc1(torch.tensor(state)))
        # Calculate the output value
        value = ____
        return value

critic_network = Critic(8)
state_value = critic_network(torch.rand(8))
print('State value:', state_value)

Modifica ed esegui il codice

Questo esercizio fa parte del corso

Deep Reinforcement Learning in Python

AvançadoNível de habilidade

4.8+

Inizia il corso gratis

Discover how deep reinforcement learning improves upon traditional Reinforcement Learning while studying and implementing your first Deep Q Learning algorithm.

Exercise 1: Introduction to deep reinforcement learning Exercise 2: Environment and neural network setup Exercise 3: DRL training loop Exercise 4: Introduction to deep Q learning Exercise 5: Deep learning and DQN Exercise 6: The Q-Network architecture Exercise 7: Instantiating the Q-Network Exercise 8: The barebone DQN algorithm Exercise 9: Barebone DQN action selection Exercise 10: Barebone DQN loss function Exercise 11: Training the barebone DQN

Dive into Deep Q-learning by implementing the original DQN algorithm, featuring Experience Replay, epsilon-greediness and fixed Q-targets. Beyond DQN, you will then explore two fascinating extensions that improve the performance and stability of Deep Q-learning: Double DQN and Prioritized Experience Replay.

Exercise 1: DQN with experience replay Exercise 2: The double-ended queue Exercise 3: Experience replay buffer Exercise 4: DQN with experience replay Exercise 5: The complete DQN algorithm Exercise 6: Epsilon-greediness Exercise 7: Fixed Q-targets Exercise 8: Implementing the complete DQN algorithm Exercise 9: Double DQN Exercise 10: Online network and target network in DDQN Exercise 11: Training the double DQN Exercise 12: Prioritized experience replay Exercise 13: Prioritized experience replay buffer Exercise 14: Sampling from the PER buffer Exercise 15: DQN with prioritized experience replay

Learn about the foundational concepts of policy gradient methods found in DRL. You will begin with the policy gradient theorem, which forms the basis for these methods. Then, you will implement the REINFORCE algorithm, a powerful approach to learning policies. The chapter will then guide you through Actor-Critic methods, focusing on the Advantage Actor-Critic (A2C) algorithm, which combines the strengths of both policy gradient and value-based methods to enhance learning efficiency and stability.

Exercise 1: Introduzione ai metodi policy gradient Exercise 2: L'architettura della rete di policy Exercise 3: Lavorare con distribuzioni discrete Exercise 4: Policy gradient e REINFORCE Exercise 5: Selezione dell'azione in REINFORCE Exercise 6: Addestrare l'algoritmo REINFORCE Exercise 7: Advantage Actor-Critic Exercise 8: Rete del Critic

Esercizio in corso

Exercise 9: Calcolo delle loss di Actor-Critic Exercise 10: Addestrare l’algoritmo A2C

Explore Proximal Policy Optimization (PPO) for robust DRL performance. Next, you will examine using an entropy bonus in PPO, which encourages exploration by preventing premature convergence to deterministic policies. You'll also learn about batch updates in policy gradient methods. Finally, you will learn about hyperparameter optimization with Optuna, a powerful tool for optimizing performance in your DRL models.

Exercise 1: Proximal policy optimization Exercise 2: The clipped probability ratio Exercise 3: The clipped surrogate objective function Exercise 4: Entropy bonus and PPO Exercise 5: Entropy playground Exercise 6: Training the PPO algorithm Exercise 7: Batch updates in policy gradient Exercise 8: Minibatch and DRL Exercise 9: A2C with batch updates Exercise 10: Hyperparameter optimization with Optuna Exercise 11: Hyperparameter or not?Exercise 12: Hands-on with Optuna Exercise 13: Congratulations!