rlPPOAgent

Proximal policy optimization reinforcement learning agent

Description

Proximal policy optimization (PPO) is a model-free, online, on-policy, policy gradient reinforcement learning method. This algorithm alternates between sampling data through environmental interaction and optimizing a clipped surrogate objective function using stochastic gradient descent. The action space can be either discrete or continuous.

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

Creation

Syntax

agent = rlPPOAgent(observationInfo,actionInfo)

agent = rlPPOAgent(observationInfo,actionInfo,initOpts)

agent = rlPPOAgent(actor,critic)

agent = rlPPOAgent(___,agentOptions)

Description

Create Agent from Observation and Action Specifications

agent = rlPPOAgent(observationInfo,actionInfo) creates a proximal policy optimization (PPO) agent for an environment with the given observation and action specifications, using default initialization options. The actor and critic representations in the agent use default deep neural networks built from the observation specification observationInfo and the action specification actionInfo.

agent = rlPPOAgent(observationInfo,actionInfo,initOpts) creates a PPO agent for an environment with the given observation and action specifications. The agent uses default networks configured using options specified in the initOpts object. Actor-critic agents do not support recurrent neural networks. For more information on the initialization options, see rlAgentInitializationOptions.

Create Agent from Actor and Critic Representations

agent = rlPPOAgent(actor,critic) creates a PPO agent with the specified actor and critic, using the default options for the agent.

Specify Agent Options

agent = rlPPOAgent(___,agentOptions) creates a PPO agent and sets the AgentOptions property to the agentOptions input argument. Use this syntax after any of the input arguments in the previous syntaxes.

Input Arguments

observationInfo — Observation specifications

specification object | array of specification objects

Observation specifications, specified as a reinforcement learning specification object or an array of specification objects defining properties such as dimensions, data type, and names of the observation signals.

You can extract observationInfo from an existing environment or agent using getObservationInfo. You can also construct the specifications manually using rlFiniteSetSpec or rlNumericSpec.

actionInfo — Action specification

specification object

Action specifications, specified as a reinforcement learning specification object defining properties such as dimensions, data type, and names of the action signals.

For a discrete action space, you must specify actionInfo as an rlFiniteSetSpec object.

For a continuous action space, you must specify actionInfo as an rlNumericSpec object.

You can extract actionInfo from an existing environment or agent using getActionInfo. You can also construct the specification manually using rlFiniteSetSpec or rlNumericSpec.

initOpts — Agent initialization options

rlAgentInitializationOptions object

Agent initialization options, specified as an rlAgentInitializationOptions object.

actor — Actor network representation

rlStochasticActorRepresentation object

Actor network representation for the policy, specified as an rlStochasticActorRepresentation object. For more information on creating actor representations, see Create Policy and Value Function Representations.

Your actor representation can use a recurrent neural network as its function approximator. In this case, your critic must also use a recurrent neural network. For an example, see Create PPO Agent with Recurrent Neural Networks.

critic — Critic network representation

rlValueRepresentation object

Critic network representation for estimating the discounted long-term reward, specified as an rlValueRepresentation. For more information on creating critic representations, see Create Policy and Value Function Representations.

Your critic representation can use a recurrent neural network as its function approximator. In this case, your actor must also use a recurrent neural network. For an example, see Create PPO Agent with Recurrent Neural Networks.

Properties

AgentOptions — Agent options

rlPPOAgentOptions object

Agent options, specified as an rlPPOAgentOptions object.

Object Functions

train	Train reinforcement learning agents within a specified environment
sim	Simulate trained reinforcement learning agents within specified environment
getAction	Obtain action from agent or actor representation given environment observations
getActor	Get actor representation from reinforcement learning agent
setActor	Set actor representation of reinforcement learning agent
getCritic	Get critic representation from reinforcement learning agent
setCritic	Set critic representation of reinforcement learning agent
generatePolicyFunction	Create function that evaluates trained policy of reinforcement learning agent

Examples

Create Discrete PPO Agent from Observation and Action Specifications

Create an environment with a discrete action space, and obtain its observation and action specifications. For this example, load the environment used in the example Create Agent Using Deep Network Designer and Train Using Image Observations. This environment has two observations: a 50-by-50 grayscale image and a scalar (the angular velocity of the pendulum). The action is a scalar with five possible elements (a torque of either -2, -1, 0, 1, or 2 Nm applied to a swinging pole).

% load predefined environment
env = rlPredefinedEnv("SimplePendulumWithImage-Discrete");

Obtain observation and action specifications from the environment.

obsInfo = getObservationInfo(env);
actInfo = getActionInfo(env);

The agent creation function initializes the actor and critic networks randomly. You can ensure reproducibility by fixing the seed of the random generator. To do so, uncomment the following line.

% rng(0)

Create a PPO agent from the environment observation and action specifications.

agent = rlPPOAgent(obsInfo,actInfo);

To check your agent, use getAction to return the action from a random observation.

getAction(agent,{rand(obsInfo(1).Dimension),rand(obsInfo(2).Dimension)})

ans = 1x1 cell array
    {[-2]}

You can now test and train the agent within the environment.

Create Continuous PPO Agent Using Initialization Options

Create an environment with a continuous action space and obtain its observation and action specifications. For this example, load the environment used in the example Train DDPG Agent to Swing Up and Balance Pendulum with Image Observation. This environment has two observations: a 50-by-50 grayscale image and a scalar (the angular velocity of the pendulum). The action is a scalar representing a torque ranging continuously from -2 to 2 Nm.

% load predefined environment
env = rlPredefinedEnv("SimplePendulumWithImage-Continuous");

% obtain observation and action specifications
obsInfo = getObservationInfo(env);
actInfo = getActionInfo(env);

Create an agent initialization option object, specifying that each hidden fully connected layer in the network must have 128 neurons (instead of the default number, 256).

initOpts = rlAgentInitializationOptions('NumHiddenUnit',128);

The agent creation function initializes the actor and critic networks randomly. You can ensure reproducibility by fixing the seed of the random generator. To do so, uncomment the following line.

% rng(0)

Create a PPO actor-critic agent from the environment observation and action specifications.

agent = rlPPOAgent(obsInfo,actInfo,initOpts);

Reduce the critic learning rate to 1e-3.

critic = getCritic(agent);
critic.Options.LearnRate = 1e-3;
agent  = setCritic(agent,critic);

Extract the deep neural networks from both the agent actor and critic.

actorNet = getModel(getActor(agent));
criticNet = getModel(getCritic(agent));

Display the layers of the critic network, and verify that each hidden fully connected layer has 128 neurons

criticNet.Layers

ans = 
  11x1 Layer array with layers:

     1   'input_1'        Image Input       50x50x1 images
     2   'conv_1'         Convolution       64 3x3x1 convolutions with stride [1  1] and padding [0  0  0  0]
     3   'relu_input_1'   ReLU              ReLU
     4   'fc_1'           Fully Connected   128 fully connected layer
     5   'input_2'        Feature Input     1 features
     6   'fc_2'           Fully Connected   128 fully connected layer
     7   'concat'         Concatenation     Concatenation of 2 inputs along dimension 1
     8   'relu_body'      ReLU              ReLU
     9   'fc_body'        Fully Connected   128 fully connected layer
    10   'body_output'    ReLU              ReLU
    11   'output'         Fully Connected   1 fully connected layer

Plot actor and critic networks

plot(layerGraph(actorNet))

plot(layerGraph(criticNet))

To check your agent, use getAction to return the action from a random observation.

getAction(agent,{rand(obsInfo(1).Dimension),rand(obsInfo(2).Dimension)})

ans = 1x1 cell array
    {[0.9228]}

You can now test and train the agent within the environment.

Create Proximal Policy Optimization Agent

Create an environment interface, and obtain its observation and action specifications.

env = rlPredefinedEnv("CartPole-Discrete");
obsInfo = getObservationInfo(env);
actInfo = getActionInfo(env);

Create a critic representation.

% Create the network to be used as approximator in the critic.
criticNetwork = [
    featureInputLayer(4,'Normalization','none','Name','state')
    fullyConnectedLayer(1,'Name','CriticFC')];

% Set options for the critic.
criticOpts = rlRepresentationOptions('LearnRate',8e-3,'GradientThreshold',1);

% Create the critic.
critic = rlValueRepresentation(criticNetwork,obsInfo,'Observation',{'state'},criticOpts);

Create an actor representation.

% Create the network to be used as approximator in the actor.
actorNetwork = [
    featureInputLayer(4,'Normalization','none','Name','state')
    fullyConnectedLayer(2,'Name','action')];

% Set options for the actor.
actorOpts = rlRepresentationOptions('LearnRate',8e-3,'GradientThreshold',1);

% Create the actor.
actor = rlStochasticActorRepresentation(actorNetwork,obsInfo,actInfo,...
    'Observation',{'state'},actorOpts);

Specify agent options, and create a PPO agent using the environment, actor, and critic.

agentOpts = rlPPOAgentOptions(...
    'ExperienceHorizon',1024, ...
    'DiscountFactor',0.95);
agent = rlPPOAgent(actor,critic,agentOpts)

agent = 
  rlPPOAgent with properties:

    AgentOptions: [1x1 rl.option.rlPPOAgentOptions]

To check your agent, use getAction to return the action from a random observation.

getAction(agent,{rand(4,1)})

ans = 1x1 cell array
    {[-10]}

You can now test and train the agent against the environment.

Create Continuous PPO Agent

Create an environment with a continuous action space, and obtain its observation and action specifications. For this example, load the double integrator continuous action space environment used in the example Train DDPG Agent to Control Double Integrator System. The observation from the environment is a vector containing the position and velocity of a mass. The action is a scalar representing a force, applied to the mass, ranging continuously from -2 to 2 Newton.

env = rlPredefinedEnv("DoubleIntegrator-Continuous");
obsInfo = getObservationInfo(env)

obsInfo = 
  rlNumericSpec with properties:

     LowerLimit: -Inf
     UpperLimit: Inf
           Name: "states"
    Description: "x, dx"
      Dimension: [2 1]
       DataType: "double"

actInfo = getActionInfo(env)

actInfo = 
  rlNumericSpec with properties:

     LowerLimit: -Inf
     UpperLimit: Inf
           Name: "force"
    Description: [0x0 string]
      Dimension: [1 1]
       DataType: "double"

Since the action must be contained in a limited range, set the upper and lower limit of the action signal accordingly. You must do so when the network representation for the actor has a nonlinear output layer that must be scaled to produce an output in the desired range.

actInfo.LowerLimit=-2;
actInfo.UpperLimit=2;

Create a critic representation. PPO agents use a rlValueRepresentation for the critic. For continuous observation spaces, you can use either a deep neural network or a custom basis representation. For this example, create a deep neural network as the underlying approximator.

% create the network to be used as approximator in the critic
% it must take the observation signal as input and produce a scalar value
criticNet = [
    imageInputLayer([obsInfo.Dimension 1],'Normalization','none','Name','state')
    fullyConnectedLayer(10,'Name', 'fc_in')
    reluLayer('Name', 'relu')
    fullyConnectedLayer(1,'Name','out')];

% set some training options for the critic
criticOpts = rlRepresentationOptions('LearnRate',8e-3,'GradientThreshold',1);

% create the critic representation from the network
critic = rlValueRepresentation(criticNet,obsInfo,'Observation',{'state'},criticOpts);

PPO agents use a rlStochasticActorRepresentation. For continuous action spaces, stochastic actors can only use a neural network approximator.

The observation input (here called myobs) must accept a two-dimensional vector, as specified in obsInfo. The output (here called myact) must also be a two-dimensional vector (twice the number of dimensions specified in actInfo). The elements of the output vector represent, in sequence, all the means and all the standard deviations of every action (in this case there is only one mean value and one standard deviation).

The fact that standard deviations must be non-negative while mean values must fall within the output range means that the network must have two separate paths. The first path is for the mean values, and any output nonlinearity must be scaled so that it can produce outputs in the output range. The second path is for the standard deviations, and you must use a softplus or relu layer to enforce non-negativity.

% input path layers (2 by 1 input and a 1 by 1 output)
inPath = [ 
    imageInputLayer([obsInfo.Dimension 1], 'Normalization','none','Name','state')
    fullyConnectedLayer(10,'Name', 'ip_fc')   % 10 by 1 output
    reluLayer('Name', 'ip_relu')              % nonlinearity
    fullyConnectedLayer(1,'Name','ip_out') ]; % 1 by 1 output

% path layers for mean value (1 by 1 input and 1 by 1 output)
% using scalingLayer to scale the range
meanPath = [
    fullyConnectedLayer(15,'Name', 'mp_fc1') % 15 by 1 output
    reluLayer('Name', 'mp_relu')             % nonlinearity
    fullyConnectedLayer(1,'Name','mp_fc2');  % 1 by 1 output
    tanhLayer('Name','tanh');                % output range: (-1,1)
    scalingLayer('Name','mp_out','Scale',actInfo.UpperLimit) ]; % output range: (-2N,2N)

% path layers for standard deviation (1 by 1 input and output)
% using softplus layer to make it non negative
sdevPath = [
    fullyConnectedLayer(15,'Name', 'vp_fc1') % 15 by 1 output
    reluLayer('Name', 'vp_relu')             % nonlinearity
    fullyConnectedLayer(1,'Name','vp_fc2');  % 1 by 1 output
    softplusLayer('Name', 'vp_out') ];       % output range: (0,+Inf)

% conctatenate two inputs (along dimension #3) to form a single (2 by 1) output layer
outLayer = concatenationLayer(1,2,'Name','mean&sdev');

% add layers to layerGraph network object
actorNet = layerGraph(inPath);
actorNet = addLayers(actorNet,meanPath);
actorNet = addLayers(actorNet,sdevPath);
actorNet = addLayers(actorNet,outLayer);

% connect layers: you must connect the mean value path to the first input of the concatenation layer
actorNet = connectLayers(actorNet,'ip_out','mp_fc1/in');     % connect output of inPath to meanPath input
actorNet = connectLayers(actorNet,'ip_out','vp_fc1/in');     % connect output of inPath to sdevPath input
actorNet = connectLayers(actorNet,'mp_out','mean&sdev/in1'); % connect output of meanPath to mean&sdev input #1
actorNet = connectLayers(actorNet,'vp_out','mean&sdev/in2'); % connect output of sdevPath to mean&sdev input #2

% plot network 
plot(actorNet)

Specify some options for the actor and create the stochastic actor representation using the deep neural network actorNet.

% set some training options for the actor
actorOpts = rlRepresentationOptions('LearnRate',8e-3,'GradientThreshold',1);

% create the actor using the network
actor = rlStochasticActorRepresentation(actorNet,obsInfo,actInfo,...
    'Observation',{'state'},actorOpts);

Specify agent options, and create a PPO agent using the actor, critic and agent options.

agentOpts = rlPPOAgentOptions(...
    'ExperienceHorizon',1024, ...
    'DiscountFactor',0.95);
agent = rlPPOAgent(actor,critic,agentOpts)

agent = 
  rlPPOAgent with properties:

    AgentOptions: [1x1 rl.option.rlPPOAgentOptions]

To check your agent, use getAction to return the action from a random observation.

getAction(agent,{rand(2,1)})

ans = 1x1 cell array
    {[0.6668]}

You can now test and train the agent within the environment.

Create PPO Agent with Recurrent Neural Networks

For this example load the predefined environment used for the Train DQN Agent to Balance Cart-Pole System example.

env = rlPredefinedEnv('CartPole-Discrete');

Get observation and action information. This environment has a continuous four-dimensional observation space (the positions and velocities of both cart and pole) and a discrete one-dimensional action space consisting on the application of two possible forces, -10N or 10N.

obsInfo = getObservationInfo(env);
actInfo = getActionInfo(env);

Create a recurrent deep neural network for the critic. To create a recurrent neural network, use a sequenceInputLayer as the input layer and include an lstmLayer as one of the other network layers.

criticNetwork = [
    sequenceInputLayer(obsInfo.Dimension(1),'Normalization','none','Name','myobs')
    fullyConnectedLayer(8, 'Name', 'fc')
    reluLayer('Name','relu')
    lstmLayer(8,'OutputMode','sequence','Name','lstm')
    fullyConnectedLayer(1,'Name','output')];

Create a value function representation object for the critic.

criticOptions = rlRepresentationOptions('LearnRate',1e-2,'GradientThreshold',1);
critic = rlValueRepresentation(criticNetwork,obsInfo,...
    'Observation','myobs', criticOptions);

Define a recurrent neural network for the actor. Since the critic has a recurrent network, the actor must have a recurrent network too.

actorNetwork = [
    sequenceInputLayer(obsInfo.Dimension(1),'Normalization','none','Name','myobs')
    fullyConnectedLayer(8,'Name','fc')
    reluLayer('Name','relu')
    lstmLayer(8,'OutputMode','sequence','Name','lstm')
    fullyConnectedLayer(numel(actInfo.Elements),'Name','output')
    softmaxLayer('Name','actionProb')];

Create a stochastic actor representation for the network.

actorOptions = rlRepresentationOptions('LearnRate',1e-3,'GradientThreshold',1);
actor = rlStochasticActorRepresentation(actorNetwork,obsInfo,actInfo,...
    'Observation','myobs', actorOptions);

Create the agent options object.

agentOptions = rlPPOAgentOptions(...
    'AdvantageEstimateMethod', 'finite-horizon', ...
    'ClipFactor', 0.1);

When recurrent neural networks are used, the MiniBatchSize property is the length of the learning trajectory.

agentOptions.MiniBatchSize

ans = 128

Create the agent using the actor and critic representations, as well as the agent options object.

agent = rlPPOAgent(actor,critic,agentOptions);

Check your agent using getAction to return the action from a random observation.

getAction(agent,rand(obsInfo.Dimension))

ans = 1x1 cell array
    {[10]}

Tips

• For continuous action spaces, this agent does not enforce the constraints set by the action specification. In this case, you must enforce action space constraints within the environment.

• While tuning the learning rate of the actor network is necessary for PPO agents, it is not necessary for TRPO agents.