L'architecture du réseau politique

Construisez l'architecture d'un réseau de politiques que vous pourrez utiliser ultérieurement pour former votre agent de gradient de politique.

Le réseau de politiques prend l'état comme entrée et produit une probabilité dans l'espace d'action. Dans l'environnement Lunar Lander, vous travaillez avec quatre actions distinctes. Vous souhaitez donc que votre réseau génère une probabilité pour chacune de ces actions.

Cet exercice fait partie du cours

<cours>Apprentissage par renforcement profond en Python</cours>

Instructions de l’exercice

Indiquez la taille de la couche de sortie du réseau de politiques ; pour plus de flexibilité, utilisez le nom de la variable plutôt que le nombre réel.
Veuillez vous assurer que la couche finale renvoie des probabilités.

Exercice interactif pratique

Essayez cet exercice en complétant ce code d’exemple.

class PolicyNetwork(nn.Module):
  def __init__(self, state_size, action_size):
    super(PolicyNetwork, self).__init__()
    self.fc1 = nn.Linear(state_size, 64)
    self.fc2 = nn.Linear(64, 64)
    # Give the desired size for the output layer
    self.fc3 = nn.Linear(64, ____)

  def forward(self, state):
    x = torch.relu(self.fc1(torch.tensor(state)))
    x = torch.relu(self.fc2(x))
    # Obtain the action probabilities
    action_probs = ____(self.fc3(x), dim=-1)
    return action_probs
  
policy_network = PolicyNetwork(8, 4)
action_probs = policy_network(state)
print('Action probabilities:', action_probs)

Modifier et exécuter le code

Cet exercice fait partie du cours

<cours>Apprentissage par renforcement profond en Python</cours>

AvancéNiveau de compétence

4.8+

Commencer le cours gratuitement

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: Introduction au gradient de politique Exercise 2: L'architecture du réseau politique

Exercice actuel

Exercise 3: Utilisation des distributions discrètes Exercise 4: Gradient de politique et REINFORCE Exercise 5: Sélection d'actions dans REINFORCE Exercise 6: Formation de l'algorithme REINFORCE Exercise 7: Critique de l'acteur Exercise 8: Réseau de critiques Exercise 9: Calcul des pertes pour les acteurs critiques Exercise 10: Formation de l'algorithme 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!