"""
Independent Q-learning (IQL)
Implementation: TensorFlow 2.X
"""
from argparse import Namespace
from xuance.common import List
from xuance.tensorflow import tf, tk, Module
from xuance.tensorflow.learners import LearnerMAS
[docs]
class IQL_Learner(LearnerMAS):
def __init__(self,
config: Namespace,
model_keys: List[str],
agent_keys: List[str],
policy: Module,
callback):
super(IQL_Learner, self).__init__(config, model_keys, agent_keys, policy, callback)
self.build_optimizer()
self.gamma = config.gamma
self.sync_frequency = config.sync_frequency
self.n_actions = {k: self.policy.action_space[k].n for k in self.model_keys}
[docs]
def build_optimizer(self):
if ("macOS" in self.os_name) and ("arm" in self.os_name): # For macOS with Apple's M-series chips.
self.optimizer = {k: tk.optimizers.legacy.Adam(self.config.learning_rate) for k in self.model_keys}
else:
self.optimizer = {k: tk.optimizers.Adam(self.config.learning_rate) for k in self.model_keys}
@tf.function
def forward_fn(self, *args):
bs, obs, actions, rewards, obs_next, terminals, agent_mask, avail_actions, avail_actions_next, IDs = args
info_train, gradients = {}, {}
for key in self.model_keys:
with tf.GradientTape() as tape:
mask_values = agent_mask[key]
_, _, q_eval = self.policy(observation=obs, agent_ids=IDs, avail_actions=avail_actions, agent_key=key)
_, q_next = self.policy.Qtarget(observation=obs_next, agent_ids=IDs, agent_key=key)
q_eval_a = tf.gather(q_eval[key], tf.cast(actions[key][:, None], dtype=tf.int32),
axis=-1, batch_dims=-1)
q_eval_a = tf.reshape(q_eval_a, [bs])
if self.use_actions_mask:
q_next[key][avail_actions_next[key] == 0] = -1e10
if self.config.double_q:
_, actions_next_greedy, _ = self.policy(obs_next, IDs, agent_key=key, avail_actions=avail_actions)
q_next_a = tf.gather(q_next[key], actions_next_greedy[key][:, None], axis=-1, batch_dims=-1)
q_next_a = tf.reshape(q_next_a, [bs])
else:
q_next_a = q_next[key].max(dim=-1, keepdim=True).values.reshape(bs)
q_target = rewards[key] + (1 - terminals[key]) * self.gamma * q_next_a
# calculate the loss function
td_error = (q_eval_a - tf.stop_gradient(q_target)) * mask_values
loss = tf.reduce_sum(td_error ** 2) / tf.reduce_sum(mask_values)
gradients[key] = tape.gradient(loss, self.policy.parameters_model(key))
if self.use_grad_clip:
gradients[key], _ = tf.clip_by_global_norm(gradients[key], clip_norm=self.grad_clip_norm)
self.optimizer[key].apply_gradients(zip(gradients[key], self.policy.parameters_model(key)))
else:
self.optimizer[key].apply_gradients(zip(gradients[key], self.policy.parameters_model(key)))
info_train.update({
f"{key}/loss_Q": loss,
f"{key}/predictQ": tf.reduce_mean(q_eval_a)
})
info_train.update(self.callback.on_update_agent_wise(self.iterations, key,
info=info_train, method="update",
mask_values=mask_values, q_eval_a=q_eval_a,
q_next_a=q_next_a, q_target=q_target,
td_error=td_error, loss=loss))
return info_train
@tf.function
def learn(self, *inputs):
if self.distributed_training:
info_train = self.policy.mirrored_strategy.run(self.forward_fn, args=inputs)
return info_train[0]
else:
return self.forward_fn(*inputs)
[docs]
def update(self, sample):
self.iterations += 1
# prepare training data
sample_Tensor = self.build_training_data(sample=sample,
use_parameter_sharing=self.use_parameter_sharing,
use_actions_mask=self.use_actions_mask)
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']
avail_actions = sample_Tensor['avail_actions']
avail_actions_next = sample_Tensor['avail_actions_next']
IDs = sample_Tensor['agent_ids']
if self.use_parameter_sharing:
key = self.model_keys[0]
bs = batch_size * self.n_agents
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
info = self.callback.on_update_start(self.iterations, method="update",
policy=self.policy, sample_Tensor=sample_Tensor, bs=bs)
info_train = self.learn(bs, obs, actions, rewards, obs_next, terminals,
agent_mask, avail_actions, avail_actions_next, IDs)
for k, v in info_train.items():
info_train[k] = v.numpy()
info.update(info_train)
if self.iterations % self.sync_frequency == 0:
self.policy.copy_target()
info.update(self.callback.on_update_end(self.iterations, method="update_rnn", policy=self.policy, info=info))
return info