Last updated
Last updated
Actor-Critic algorithms combine value function and policy estimation. They consist of an actor, which learns a parameterized policy, , mapping states to probability distributions over actions, and a critic, which learns a parameterized function which assigns a value to actions taken.
There are a variety of approaches to training the critic. Three options are provided with the baseline algorithms.
Learn the V function and use it to approximate the Q function. See for some examples specs.
Advantage with n-step forward returns from . See for some example specs.
Generalized advantage estimation from . See for some example specs.
The actor and critic can be trained separately or jointly, depending on whether the networks are structured to share parameters.
The critic is trained using temporal difference learning, similarly to the algorithm. The actor is trained using policy gradients, similarly to . However in this case, the estimate of the "goodness" of actions is evaluated using the advantage, , which is calculated using the critic's estimation of . measures how much better the action taken in states s
was compared with expected value of being in that state s
, i.e. the expectation over all possible actions in state s
.
Actor-Critic algorithms are on policy. Only data that is gathered using the current policy can be used to update the parameters. Once the policy parameters have been updated all previous data gathered must be discarded and the collection process started again with the new policy.
Algorithm: Simple Actor Critic with separate actor and critic parameters
See and for example specs of variations of the Actor-Critic algorithm.
Basic Parameters
algorithm
action_policy
string specifying which policy to use to act. "Categorical" (for discrete action spaces), "Normal" (for continuous actions spaces with one dimension), or "default" to automatically switch between the two depending on the environment.
use_gae
whether to calculate the advantage using generalized advantage estimation.
use_nstep
: whether to calculate the advantage using n-step forward returns. If use_gae
is true
this parameter will be ignored.
num_step_returns
number of forward steps to use when calculating the target for advantage estimation using nstep forward returns.
add_entropy
whether to add entropy to the advantage to encourage exploration
entropy_coef
coefficient to multiply the entropy of the distribution with when adding it to advantage
training_frequency
when using episodic data storage (memory) such as "OnPolicyReplay", this means how many episodes of data to collect before each training iteration - a common value is 1; or when using batch data storage (memory) such as "OnPolicyBatchReplay", how often to train the algorithm. Value of 32 means train every 32 steps the agent takes in the environment using the 32 examples since the agent was previously trained.
memory
net
shared
whether the actor and the critic should share parameters
Advanced Parameters
algorithm
training_epoch
how many gradient steps to take when training the critic. Only applies when the actor and critic have separate parameters.
policy_loss_coef
how much weight to give to the policy (actor) component of the loss when the actor and critic have shared parameters, so are trained jointly.
val_loss_coef
how much weight to give to the critic component of the loss when the actor and critic have shared parameters, so are trained jointly.
net
use_same_optim
whether to use the optim_actor
for both the actor and critic. This can be useful when using conducting a parameter search.
name
action_pdtype
gamma
lam
trade off between bias and variance when using generalized advantage estimation. 0 corresponds to low variance and high bias. 1 corresponds to high variance and low bias.
name
. Compatible types;
type
. All are compatible.
hid_layers
hid_layers_activation
actor_optim_spec
optimizer for the actor
critic_optim_spec
optimizer for the critic
rnn_hidden_size
rnn_num_layers
seq_len
clip_grad
:
clip_grad_val
:
lr_decay
:
lr_decay_frequency
:
lr_decay_min_timestep
:
lr_anneal_timestep
:
gpu
: