"""
Multi-agent Soft Actor-critic (MASAC)
Implementation: TensorFlow 2.X
"""
from argparse import Namespace
from operator import itemgetter
from xuance.common import List
from xuance.tensorflow import tf, Module
from xuance.tensorflow.learners.multi_agent_rl.isac_learner import ISAC_Learner
[docs]
class MASAC_Learner(ISAC_Learner):
def __init__(self,
config: Namespace,
model_keys: List[str],
agent_keys: List[str],
policy: Module,
callback):
super(MASAC_Learner, self).__init__(config, model_keys, agent_keys, policy, callback)
# @tf.function
[docs]
def forward_fn(self, *args):
batch_size, bs, obs, obs_joint, actions, actions_joint, rewards, obs_next, next_obs_joint, terminals, IDs, agent_mask = args
info_train, gradients_c, gradients_a, gradients_alpha = {}, {}, {}, {}
with tf.GradientTape(persistent=True) as tape:
# Update critic
_, actions_next, log_pi_next = self.policy(observation=obs_next, agent_ids=IDs)
if self.use_parameter_sharing:
key = self.model_keys[0]
actions_next_joint = tf.reshape(tf.reshape(actions_next[key], [batch_size, self.n_agents, -1]),
[batch_size, -1])
else:
actions_next_joint = tf.reshape(tf.stack(itemgetter(*self.model_keys)(actions_next), -1),
[batch_size, -1])
_, _, action_q_1, action_q_2 = self.policy.Qpolicy(joint_observation=obs_joint, joint_actions=actions_joint,
agent_ids=IDs)
_, _, target_q = self.policy.Qtarget(joint_observation=next_obs_joint, joint_actions=actions_next_joint,
agent_ids=IDs)
for key in self.model_keys:
mask_values = agent_mask[key]
action_q_1_i, action_q_2_i = tf.reshape(action_q_1[key], [bs]), tf.reshape(action_q_2[key], [bs])
log_pi_next_eval = tf.reshape(log_pi_next[key], [bs])
target_value = tf.reshape(target_q[key], [bs]) - self.alpha[key] * log_pi_next_eval
backup = rewards[key] + (1 - terminals[key]) * self.gamma * target_value
backup = tf.stop_gradient(backup)
td_error_1, td_error_2 = action_q_1_i - backup, action_q_2_i - backup
td_error_1 *= mask_values
td_error_2 *= mask_values
loss_c = (tf.reduce_sum(td_error_1 ** 2) + tf.reduce_sum(td_error_2 ** 2)) / tf.reduce_sum(mask_values)
gradients_c[key] = tape.gradient(loss_c, self.policy.critic_trainable_variables(key))
if self.use_grad_clip:
gradients_c[key], _ = tf.clip_by_global_norm(gradients_c[key], clip_norm=self.grad_clip_norm)
self.optimizer[key]['critic'].apply_gradients(zip(gradients_c[key],
self.policy.critic_trainable_variables(key)))
else:
self.optimizer[key]['critic'].apply_gradients(zip(gradients_c[key],
self.policy.critic_trainable_variables(key)))
info_train.update({f"{key}/loss_critic": loss_c})
# Update actor
_, actions_eval, log_pi_eval = self.policy(observation=obs, agent_ids=IDs)
log_pi_eval_i = {}
for key in self.model_keys:
mask_values = agent_mask[key]
if self.use_parameter_sharing:
actions_eval_joint = tf.reshape(tf.reshape(actions_eval[key], [batch_size, self.n_agents, -1]),
[batch_size, -1])
else:
a_joint = {k: actions_eval[k] if k == key else actions[k] for k in self.agent_keys}
actions_eval_joint = tf.reshape(tf.stack(itemgetter(*self.model_keys)(a_joint),
axis=-1), [batch_size, -1])
_, _, policy_q_1, policy_q_2 = self.policy.Qpolicy(joint_observation=obs_joint,
joint_actions=actions_eval_joint,
agent_ids=IDs, agent_key=key)
log_pi_eval_i[key] = tf.reshape(log_pi_eval[key], [bs])
policy_q = tf.reshape(tf.math.minimum(policy_q_1[key], policy_q_2[key]), [bs])
loss_a = tf.reduce_sum((self.alpha[key] * log_pi_eval_i[key] - policy_q) * mask_values) / tf.reduce_sum(
mask_values)
gradients_a[key] = tape.gradient(loss_a, self.policy.actor_trainable_variables(key))
if self.use_grad_clip:
gradients_a[key], _ = tf.clip_by_global_norm(gradients_a[key], clip_norm=self.grad_clip_norm)
self.optimizer[key]['actor'].apply_gradients(zip(gradients_a[key],
self.policy.actor_trainable_variables(key)))
else:
self.optimizer[key]['actor'].apply_gradients(zip(gradients_a[key],
self.policy.actor_trainable_variables(key)))
info_train.update({f"{key}/loss_actor": loss_a,
f"{key}/predictQ": tf.math.reduce_mean(policy_q)})
# Automatically entropy tuning
if self.use_automatic_entropy_tuning:
for key in self.model_keys:
alpha_loss = -tf.math.reduce_mean(
self.alpha_layer[key].log_alpha.value() * (log_pi_eval_i[key] + self.target_entropy[key]))
gradients_alpha[key] = tape.gradient(alpha_loss, self.alpha_layer[key].trainable_variables)
gradients_alpha[key], _ = tf.clip_by_global_norm(gradients_alpha[key],
clip_norm=self.grad_clip_norm)
self.alpha_optimizer[key].apply_gradients(zip(gradients_alpha[key],
self.alpha_layer[key].trainable_variables))
self.alpha[key] = tf.math.exp(self.alpha_layer[key].log_alpha)
info_train.update({f"{key}/alpha_loss": alpha_loss,
f"{key}/alpha": self.alpha[key]})
else:
for key in self.model_keys:
info_train.update({f"{key}/alpha_loss": tf.Tensor(0.0, dtype=tf.float32),
f"{key}/alpha": self.alpha[key]})
return info_train
[docs]
def update(self, sample):
self.iterations += 1
# Prepare training data.
sample_Tensor = self.build_training_data(sample,
use_parameter_sharing=self.use_parameter_sharing,
use_actions_mask=False)
batch_size = sample_Tensor['batch_size']
obs = sample_Tensor['obs']
actions = sample_Tensor['actions']
obs_next = sample_Tensor['obs_next']
rewards = sample_Tensor['rewards']
terminals = sample_Tensor['terminals']
agent_mask = sample_Tensor['agent_mask']
IDs = sample_Tensor['agent_ids']
if self.use_parameter_sharing:
key = self.model_keys[0]
bs = batch_size * self.n_agents
obs_joint = tf.reshape(obs[key], [batch_size, -1])
next_obs_joint = tf.reshape(obs_next[key], [batch_size, -1])
actions_joint = tf.reshape(actions[key], [batch_size, -1])
rewards[key] = tf.reshape(rewards[key], [batch_size * self.n_agents])
terminals[key] = tf.reshape(terminals[key], [batch_size * self.n_agents])
else:
bs = batch_size
obs_joint = tf.reshape(tf.stack(itemgetter(*self.agent_keys)(obs), axis=-1), [batch_size, -1])
next_obs_joint = tf.reshape(tf.stack(itemgetter(*self.agent_keys)(obs_next), axis=-1), [batch_size, -1])
actions_joint = tf.reshape(tf.stack(itemgetter(*self.agent_keys)(actions), axis=-1), [batch_size, -1])
info = self.callback.on_update_start(self.iterations, method="update", policy=self.policy,
sample_Tensor=sample_Tensor, bs=bs, obs_joint=obs_joint,
next_obs_joint=next_obs_joint, actions_joint=actions_joint)
info_train = self.learn(batch_size, bs, obs, obs_joint, actions, actions_joint, rewards,
obs_next, next_obs_joint, terminals, IDs, agent_mask)
for k, v in info_train.items():
info_train[k] = v.numpy()
info.update(info_train)
self.policy.soft_update(self.tau)
info.update(self.callback.on_update_end(self.iterations, method="update", policy=self.policy, info=info))
return info