"""
Weighted QMIX
Paper link:
https://proceedings.neurips.cc/paper/2020/file/73a427badebe0e32caa2e1fc7530b7f3-Paper.pdf
Implementation: TensorFlow 2.X
"""
import numpy as np
from argparse import Namespace
from operator import itemgetter
from xuance.common import List
from xuance.tensorflow import tf, tk, Module
from xuance.tensorflow.learners import LearnerMAS
[docs]
class WQMIX_Learner(LearnerMAS):
def __init__(self,
config: Namespace,
model_keys: List[str],
agent_keys: List[str],
policy: Module,
callback):
super(WQMIX_Learner, self).__init__(config, model_keys, agent_keys, policy, callback)
self.build_optimizer()
self.alpha = config.alpha
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}
self.mse_loss = tk.losses.MeanSquaredError()
[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 = tk.optimizers.legacy.Adam(self.config.learning_rate)
else:
self.optimizer = tk.optimizers.Adam(self.config.learning_rate)
@tf.function
def forward_fn(self, bs, batch_size, state, obs, actions, rewards_tot, state_next, obs_next, terminals_tot,
agent_mask, avail_actions, avail_actions_next, IDs):
with tf.GradientTape() as tape:
# calculate Q_tot
_, action_max, q_eval = self.policy(observation=obs, agent_ids=IDs, avail_actions=avail_actions)
_, q_eval_centralized = self.policy.q_centralized(observation=obs, agent_ids=IDs)
_, q_eval_next_centralized = self.policy.target_q_centralized(observation=obs_next, agent_ids=IDs)
q_eval_a, q_eval_centralized_a, q_eval_next_centralized_a, act_next = {}, {}, {}, {}
for key in self.model_keys:
actions_eval = tf.cast(actions[key][:, None], dtype=tf.int32)
q_eval_a[key] = tf.reshape(tf.gather(q_eval[key], actions_eval, axis=-1, batch_dims=-1), [bs])
q_eval_centralized_a[key] = tf.reshape(tf.gather(q_eval_centralized[key], actions_eval,
axis=-1, batch_dims=-1), [bs])
if self.config.double_q:
_, a_next_greedy, _ = self.policy(observation=obs_next, agent_ids=IDs,
avail_actions=avail_actions_next, agent_key=key)
act_next[key] = tf.expand_dims(a_next_greedy[key], axis=-1)
else:
_, q_next_eval = self.policy.Qtarget(observation=obs_next, agent_ids=IDs, agent_key=key)
if self.use_actions_mask:
q_next_eval[key][avail_actions_next[key] == 0] = -1e10
act_next[key] = tf.argmax(q_next_eval[key], axis=-1)
q_eval_next_centralized_a[key] = tf.reshape(tf.gather(q_eval_next_centralized[key], act_next[key],
axis=-1, batch_dims=-1), [bs])
q_eval_a[key] *= agent_mask[key]
q_eval_centralized_a[key] *= agent_mask[key]
q_eval_next_centralized_a[key] *= agent_mask[key]
q_tot_eval = self.policy.Q_tot(q_eval_a, state) # calculate Q_tot
q_tot_centralized = self.policy.q_feedforward(q_eval_centralized_a, state) # calculate centralized Q
q_tot_next_centralized = self.policy.target_q_feedforward(q_eval_next_centralized_a, state_next) # y_i
target_value = rewards_tot + (1 - terminals_tot) * self.gamma * q_tot_next_centralized
target_value = tf.stop_gradient(target_value)
td_error = q_tot_eval - target_value
# calculate weights
ones = tf.ones_like(td_error)
w = ones * self.alpha
if self.config.agent == "CWQMIX":
condition_1 = ((action_max == actions.reshape([-1, self.n_agents, 1])) * agent_mask).all(dim=1)
condition_2 = target_value > q_tot_centralized
conditions = condition_1 | condition_2
w = tf.where(conditions, ones, w)
elif self.config.agent == "OWQMIX":
condition = td_error < 0
w = tf.where(condition, ones, w)
else:
raise AttributeError(f"The agent named is {self.config.agent} is currently not supported.")
# calculate losses and train
target_value = tf.reshape(target_value, [batch_size])
q_tot_centralized = tf.reshape(q_tot_centralized, [batch_size])
td_error = tf.reshape(td_error, [batch_size])
w = tf.reshape(w, [batch_size])
loss_central = self.mse_loss(target_value, q_tot_centralized)
loss_qmix = tf.reduce_mean(tf.stop_gradient(w) * (td_error ** 2))
loss = loss_qmix + loss_central
gradients = tape.gradient(loss, self.policy.parameters_model)
if self.use_grad_clip:
gradients, _ = tf.clip_by_global_norm(gradients, clip_norm=self.grad_clip_norm)
self.optimizer.apply_gradients(zip(gradients, self.policy.parameters_model))
else:
self.optimizer.apply_gradients(zip(gradients, self.policy.parameters_model))
return loss_qmix, loss_central, loss, tf.math.reduce_mean(q_tot_eval)
@tf.function
def learn(self, *inputs):
if self.distributed_training:
loss_qmix, loss_central, loss, predictQ = self.policy.mirrored_strategy.run(self.forward_fn, args=inputs)
return (self.policy.mirrored_strategy.reduce(tf.distribute.ReduceOp.SUM, loss_qmix, axis=None),
self.policy.mirrored_strategy.reduce(tf.distribute.ReduceOp.SUM, loss_central, axis=None),
self.policy.mirrored_strategy.reduce(tf.distribute.ReduceOp.SUM, loss, axis=None),
self.policy.mirrored_strategy.reduce(tf.distribute.ReduceOp.SUM, predictQ, axis=None))
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,
use_global_state=True)
batch_size = sample_Tensor['batch_size']
state = sample_Tensor['state']
state_next = sample_Tensor['state_next']
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_tot = tf.reshape(tf.reduce_mean(rewards[key], axis=1), [batch_size, 1])
terminals_tot = tf.reshape(tf.reduce_prod(terminals[key], axis=1), [batch_size, 1])
else:
bs = batch_size
rewards_tot = tf.reduce_mean(tf.stack(itemgetter(*self.agent_keys)(rewards), axis=1),
axis=-1, keepdims=True)
terminals_tot = tf.reduce_prod(tf.stack(itemgetter(*self.agent_keys)(terminals), axis=1),
axis=1, keepdims=True)
info = self.callback.on_update_start(self.iterations, method="update", policy=self.policy,
sample_Tensor=sample_Tensor, bs=bs,
rewards_tot=rewards_tot, terminals_tot=terminals_tot)
loss_qmix, loss_central, loss, predictQ = self.learn(bs, batch_size, state, obs, actions, rewards_tot,
state_next, obs_next, terminals_tot,
agent_mask, avail_actions, avail_actions_next, IDs)
info.update({
"loss_Qmix": loss_qmix.numpy(),
"loss_central": loss_central.numpy(),
"loss": loss.numpy(),
"predictQ": predictQ.numpy()
})
if self.iterations % self.sync_frequency == 0:
self.policy.copy_target()
info.update(self.callback.on_update_end(self.iterations, method="update", policy=self.policy, info=info))
return info