rlSACAgentOptions

Options for SAC agent

Since R2020b

Description

Use an rlSACAgentOptions object to specify options for soft actor-critic (SAC) agents. To create a SAC agent, use rlSACAgent.

For more information, see Soft Actor-Critic (SAC) Agents.

For more information on the different types of reinforcement learning agents, see Reinforcement Learning Agents.

Creation

Syntax

opt = rlSACAgentOptions

opt = rlSACAgentOptions(Name,Value)

Description

opt = rlSACAgentOptions creates an options object for use as an argument when creating a SAC agent using all default options. You can modify the object properties using dot notation.

opt = rlSACAgentOptions(Name,Value) sets option properties using name-value pairs. For example, rlSACAgentOptions('DiscountFactor',0.95) creates an option set with a discount factor of 0.95. You can specify multiple name-value pairs. Enclose each property name in quotes.

Properties

EntropyWeightOptions — Entropy tuning options

EntropyWeightOptions object

Entropy tuning options, specified as an EntropyWeightOptions object with the following properties.

EntropyWeight — Initial entropy component weight

1 (default) | positive scalar

Initial entropy component weight, specified as a positive scalar.

LearnRate — Optimizer learning rate

3e-4 (default) | nonnegative scalar

Optimizer learning rate, specified as a nonnegative scalar. If LearnRate is zero, the EntropyWeight value is fixed during training and the TargetEntropy value is ignored.

TargetEntropy — Target entropy value

[] (default) | scalar

Target entropy value for tuning entropy weight, specified as a scalar. A higher target entropy value encourages more exploration.

If you do not specify TargetEntropy, the agent uses –A as the target value, where A is the number of actions.

Algorithm — Algorithm to tune entropy

"adam" (default) | "sgdm" | "rmsprop"

Algorithm to tune entropy, specified as one of the following strings.

• "adam" — Use the Adam optimizer. You can specify the decay rates of the gradient and squared gradient moving averages using the GradientDecayFactor and SquaredGradientDecayFactor fields of the OptimizerParameters option.

• "sgdm" — Use the stochastic gradient descent with momentum (SGDM) optimizer. You can specify the momentum value using the Momentum field of the OptimizerParameters option.

• "rmsprop" — Use the RMSProp optimizer. You can specify the decay rate of the squared gradient moving average using the SquaredGradientDecayFactor fields of the OptimizerParameters option.

For more information about these optimizers, see Stochastic Gradient Descent in Deep Learning Toolbox™.

GradientThreshold — Threshold value for gradient

Inf (default) | positive scalar

Threshold value for the entropy gradient, specified as Inf or a positive scalar. If the gradient exceeds this value, the gradient is clipped.

OptimizerParameters — Applicable parameters for optimizer

OptimizerParameters object

Applicable parameters for the optimizer, specified as an OptimizerParameters object with the following parameters. The default parameter values work well for most problems.

Parameter	Description	Default
Momentum	Contribution of previous step, specified as a scalar from 0 to 1. A value of 0 means no contribution from the previous step. A value of 1 means maximal contribution. This parameter applies only when Optimizer is "sgdm".	0.9
Epsilon	Denominator offset, specified as a positive scalar. The optimizer adds this offset to the denominator in the network parameter updates to avoid division by zero. This parameter applies only when Optimizer is "adam" or "rmsprop".	1e-8
GradientDecayFactor	Decay rate of gradient moving average, specified as a positive scalar from 0 to 1. This parameter applies only when Optimizer is "adam".	0.9
SquaredGradientDecayFactor	Decay rate of squared gradient moving average, specified as a positive scalar from 0 to 1. This parameter applies only when Optimizer is "adam" or "rmsprop".	0.999

When a particular property of OptimizerParameters is not applicable to the optimizer type specified in the Algorithm option, that property is set to "Not applicable".

To change the default values, access the properties of OptimizerParameters using dot notation.

opt = rlSACAgentOptions;
opt.EntropyWeightOptions.OptimizerParameters.GradientDecayFactor = 0.95;

UseCriticTargetEntropy — Option to use entropy in the critic targets

true (default) | false

Option to use entropy in the critic targets, specified as a either true (default, entropy is used) or false (entropy is not used). Note that this option does not affect entropy usage in the actor.

Example: UseCriticTargetEntropy=false

PolicyUpdateFrequency — Number of steps between actor policy updates

1 (default) | positive integer

Number of steps between actor policy updates, specified as a positive integer. For more information, see Training Algorithm.

CriticUpdateFrequency — Number of steps between critic updates

1 (default) | positive integer

Number of steps between critic updates, specified as a positive integer. For more information, see Training Algorithm.

NumWarmStartSteps — Number of actions to take before updating actor and critic

positive integer

Number of actions to take before updating actor and critics, specified as a positive integer. By default, the NumWarmStartSteps value is equal to the MiniBatchSize value.

NumGradientStepsPerUpdate — Number of gradient steps when updating actor and critics

1 (default) | positive integer

Number of gradient steps to take when updating actor and critics, specified as a positive integer.

ActorOptimizerOptions — Actor optimizer options

rlOptimizerOptions object

Actor optimizer options, specified as an rlOptimizerOptions object. It allows you to specify training parameters of the actor approximator such as learning rate, gradient threshold, as well as the optimizer algorithm and its parameters. For more information, see rlOptimizerOptions and rlOptimizer.

CriticOptimizerOptions — Critic optimizer options

rlOptimizerOptions object

Critic optimizer options, specified as an rlOptimizerOptions object. It allows you to specify training parameters of the critic approximator such as learning rate, gradient threshold, as well as the optimizer algorithm and its parameters. For more information, see rlOptimizerOptions and rlOptimizer.

BatchDataRegularizerOptions — Batch data regularizer options

[] (default) | rlConservativeQLearningOptions object | rlBehaviorCloningRegularizerOptions object

Batch data regularizer options, specified as an rlConservativeQLearningOptions or as an rlBehaviorCloningRegularizerOptions object. These options are typically used to train the agent offline, from existing data. If you leave this option empty, no regularizer is used.

For more information, see rlConservativeQLearningOptions and rlBehaviorCloningRegularizerOptions.

Example: opt.BatchDataRegularizerOptions = rlConservativeQLearningOptions(NumSampledActions=20)

TargetSmoothFactor — Smoothing factor for target critic updates

1e-3 (default) | positive scalar less than or equal to 1

Smoothing factor for target critic updates, specified as a positive scalar less than or equal to 1. For more information, see Target Update Methods.

TargetUpdateFrequency — Number of steps between target critic updates

1 (default) | positive integer

Number of steps between target critic updates, specified as a positive integer. For more information, see Target Update Methods.

ResetExperienceBufferBeforeTraining — Option for clearing the experience buffer

false (default) | true

Option for clearing the experience buffer before training, specified as a logical value.

SequenceLength — Maximum batch-training trajectory length when using RNN

1 (default) | positive integer

Maximum batch-training trajectory length when using a recurrent neural network, specified as a positive integer. This value must be greater than 1 when using a recurrent neural network and 1 otherwise.

MiniBatchSize — Size of random experience mini-batch

64 (default) | positive integer

Size of random experience mini-batch, specified as a positive integer. During each training episode, the agent randomly samples experiences from the experience buffer when computing gradients for updating the actor and critics. Large mini-batches reduce the variance when computing gradients but increase the computational effort.

NumStepsToLookAhead — Number of future rewards used to estimate the value of the policy

1 (default) | positive integer

Number of future rewards used to estimate the value of the policy, specified as a positive integer. For more information, see [1], Chapter 7.

ExperienceBufferLength — Experience buffer size

10000 (default) | positive integer

Experience buffer size, specified as a positive integer. During training, the agent computes updates using a mini-batch of experiences randomly sampled from the buffer.

SampleTime — Sample time of agent

1 (default) | positive scalar | -1

Sample time of agent, specified as a positive scalar or as -1. Setting this parameter to -1 allows for event-based simulations.

Within a Simulink^® environment, the RL Agent block in which the agent is specified to execute every SampleTime seconds of simulation time. If SampleTime is -1, the block inherits the sample time from its parent subsystem.

Within a MATLAB^® environment, the agent is executed every time the environment advances. In this case, SampleTime is the time interval between consecutive elements in the output experience returned by sim or train. If SampleTime is -1, the time interval between consecutive elements in the returned output experience reflects the timing of the event that triggers the agent execution.

DiscountFactor — Discount factor

0.99 (default) | positive scalar less than or equal to 1

Discount factor applied to future rewards during training, specified as a positive scalar less than or equal to 1.

Object Functions

rlSACAgent

Soft actor-critic (SAC) reinforcement learning agent

Examples

Create SAC Agent Options Object

Create a SAC agent options object, specifying the discount factor.

opt = rlSACAgentOptions(DiscountFactor=0.95)

opt = 
  rlSACAgentOptions with properties:

                   EntropyWeightOptions: [1x1 rl.option.EntropyWeightOptions]
                 UseCriticTargetEntropy: 1
                  PolicyUpdateFrequency: 1
                  CriticUpdateFrequency: 1
                      NumWarmStartSteps: 64
              NumGradientStepsPerUpdate: 1
                  ActorOptimizerOptions: [1x1 rl.option.rlOptimizerOptions]
                 CriticOptimizerOptions: [1x2 rl.option.rlOptimizerOptions]
            BatchDataRegularizerOptions: []
                     TargetSmoothFactor: 1.0000e-03
                  TargetUpdateFrequency: 1
    ResetExperienceBufferBeforeTraining: 0
                         SequenceLength: 1
                          MiniBatchSize: 64
                    NumStepsToLookAhead: 1
                 ExperienceBufferLength: 10000
                             SampleTime: 1
                         DiscountFactor: 0.9500
                             InfoToSave: [1x1 struct]

You can modify options using dot notation. For example, set the agent sample time to 0.5.

opt.SampleTime = 0.5;

For SAC agents, configure the entropy weight optimizer using the options in EntropyWeightOptions. For example, set the target entropy value to –5.

opt.EntropyWeightOptions.TargetEntropy = -5;

References

[1] Sutton, Richard S., and Andrew G. Barto. Reinforcement Learning: An Introduction. Second edition. Adaptive Computation and Machine Learning. Cambridge, Mass: The MIT Press, 2018.

Version History

Introduced in R2020b

R2022a: Simulation and deployment: UseDeterministicExploitation will be removed

The property UseDeterministicExploitation of the rlSACAgentOptions object will be removed in a future release. Use the UseExplorationPolicy property of rlSACAgent instead.

Previously, you set UseDeterministicExploitation as follows.

• Force the agent to always select the action with maximum likelihood, thereby using a greedy deterministic policy for simulation and deployment.

  agent.AgentOptions.UseDeterministicExploitation = true;

• Allow the agent to select its action by sampling its probability distribution for simulation and policy deployment, thereby using a stochastic policy that explores the observation space.

  agent.AgentOptions.UseDeterministicExploitation = false;

Starting in R2022a, set UseExplorationPolicy as follows.

• Force the agent to always select the action with maximum likelihood, thereby using a greedy deterministic policy for simulation and deployment.

  agent.UseExplorationPolicy = false;

• Allow the agent to select its action by sampling its probability distribution for simulation and policy deployment, thereby using a stochastic policy that explores the observation space.

  agent.UseExplorationPolicy = true;

Similarly to UseDeterministicExploitation, UseExplorationPolicy affects only simulation and deployment; it does not affect training.

R2022a: The default value of the ResetExperienceBufferBeforeTraining property has changed

The default value of the ResetExperienceBufferBeforeTraining has changed from true to false.

When creating a new SAC agent, if you want to clear the experience buffer before training, you must specify ResetExperienceBufferBeforeTraining as true. For example, before training, set the property using dot notation.

agent.AgentOptions.ResetExperienceBufferBeforeTraining = true;

Alternatively, you can set the property to true in an rlSACAgentOptions object and use this object to create the SAC agent.