Once you have created an environment and reinforcement learning agent, you can train the
agent in the environment using the
train function. To
configure your training, use the
function. For example, create a training option set
opt, and train agent
agent in environment
opt = rlTrainingOptions(... 'MaxEpisodes',1000,... 'MaxStepsPerEpisode',1000,... 'StopTrainingCriteria',"AverageReward",... 'StopTrainingValue',480); trainStats = train(agent,env,opt);
For more information on creating:
Agents, see Reinforcement Learning Agents
train updates the agent as training progresses. To preserve the
original agent parameters for later use, save the agent to a MAT-file.
Training terminates automatically when the conditions specified in
StopTrainingValue of your
rlTrainingOptions object are satisfied. To manually terminate training
in progress, type
ctrl-C or, in the Reinforcement Learning Episode Manager,
click Stop Training. Because
train updates the agent
at each episode, you can resume training by calling
train(agent,env,trainOpts) again, without losing the trained parameters
learned during the first call to
In general, training performs the following iterative steps:
Initialize the agent.
For each episode:
Reset the environment.
Get the initial observation s0 from the environment.
Compute the initial action a0 = μ(s0), where μ(s) is the current policy.
Set the current action to the initial action (a←a0), and set the current observation to the initial observation (s←s0).
While the episode is not finished or terminated:
Step the environment with action a to obtain the next observation s' and the reward r.
Learn from the experience set (s,a,r,s').
Compute the next action a' = μ(s').
Update the current action with the next action (a←a') and update the current observation with the next observation (s←s').
Break if the episode termination conditions defined in the environment are met.
If the training termination condition is met, terminate training. Otherwise, begin the next episode.
The specifics of how the software performs these steps depends on the configuration of the agent and environment. For instance, resetting the environment at the start of each episode can include randomizing initial state values, if you configure your environment to do so. For more information on agents and their training algorithms, see Reinforcement Learning Agents.
By default, calling the
train function opens the Reinforcement
Learning Episode Manager, which lets you visualize the progress of the training. The Episode
Manager plot shows the reward for each episode (EpisodeReward), a
running average reward value (AverageReward). Also, for agents that
have critics, plot shows the critics estimate of the discounted long-term reward at the
start of each episode (EpisodeQ0). The Episode Manager also displays
various episode and training statistics. This episode and training information is also
returned by the
For agents with a critic, Episode Q0 is the estimate of the discounted long-term reward at the start of each episode, given the initial observation of the environment. As training progresses, Episode Q0 should approach the true discounted long-term reward if the critic is well-designed, as shown in the preceding figure.
To turn off the Reinforcement Learning Episode Manager, set the
During training, you can save candidate agents that meet conditions you specify in
SaveAgentValue of your
rlTrainingOptions object. For instance, you can save any agent whose
episode reward exceeds a certain value, even if the overall condition for terminating
training is not yet satisfied. For example, to save agents when the episode reward is
opt = rlTrainingOptions('SaveAgentCriteria',"EpisodeReward",'SaveAgentValue',100');
train stores saved agents in a MAT-file in the folder you specify
SaveAgentDirectory option of
rlTrainingOptions. Saved agents can be useful, for instance, to allow
you to test candidate agents generated during a long-running training process. For details
about saving criteria and saving location, see
After training is complete, you can save the final trained agent from the MATLAB® workspace using the
save function. For example, save the
myAgent to the file
finalAgent.mat in the
current working directory.
save(opt.SaveAgentDirectory + "/finalAgent.mat",'agent')
By default, when DDPG and DQN agents are saved, the experience buffer data is not saved.
If you plan to further train your saved agent, you can start training with the previous
experience buffer as a starting point. In this case, set the
SaveExperienceBufferWithAgent agent option to
For some agents, such as those with large experience buffers and image-based observations,
the memory required for saving their experience buffer is large. In these cases, you must
ensure that there is enough memory available for the saved agents.
You can accelerate agent training by running parallel training simulations. If you have:
Parallel Computing Toolbox™ software, you can run parallel simulations on multicore computers
MATLAB Parallel Server™software, you can run parallel simulations on computer clusters or cloud resources
When training with parallel computing, the host client sends copies of the agent and environment to each parallel worker. Each worker simulates the agent within the environment and sends their simulation data back to the host. The host agent learns from the data sent by the workers and sends the updated policy parameters back to the workers.
To create a parallel pool of
N workers, type:
pool = parpool(N);
If you do not create a parallel pool using
train function automatically creates one
using your default parallel pool preferences. For more information on specifying these
preferences, see Specify Your Parallel Preferences (Parallel Computing Toolbox).
For off-policy agents, such as DDPG and DQN, do not use all of your cores for parallel training. For example, if your CPU has six cores, train with four workers. Doing so provides more resources for the host client to compute gradients based on the experiences sent back from the workers. Limiting the number of workers is not necessary for on-policy agents, such as PG and AC, when the gradients are computed on the workers.
For more information on configuring your training to use parallel computing, see
To benefit from parallel computing, the computational cost for simulating the environment must be relatively expensive compared to the optimization of parameters when sending experiences back to the host. If the simulation of the environment is not expensive enough, the workers idle while waiting for the host to learn and send back updated parameters.
When sending experiences back from the workers, you can improve sample efficiency when the ratio R = (complexity of environment step)/(complexity of learning) is large. If the environment is fast to simulate (R is small), you are unlikely to get any benefit from experience-based parallelization. If the environment is expensive to simulate but it is also expensive to learn (for example, if the mini-batch size is large) then you are also unlikely to improve sample efficiency. However in this case, for off-policy agents, you can reduce the mini-batch size to make R larger, which improves sample efficiency.
For an example that trains an agent using parallel computing in:
When using deep neural network function approximators for your actor or critic
representations, you can speed up training by performing representation operations on a GPU
rather than a CPU. To do so, set the
UseDevice option to
opt = rlRepresentationOptions('UseDevice',"gpu");
The size of any performance improvement depends on your specific application and network configuration.
When validating your agent, consider checking how your agent handles:
Changes to simulation initial conditions. To change the model initial conditions, modify the reset function for the environment. For example reset functions, see:
Mismatches between the training and simulation environment dynamics. To do so, create test environments in the same way that you created the training environment, modifying the environment behavior.
As with parallel training, if you have Parallel
Computing Toolbox software, you can run multiple parallel simulations on multicore computers. If
you have MATLAB
Parallel Server software, you can run multiple parallel simulations on computer clusters or
cloud resources. For more information on configuring your simulation to use parallel
If your training environment implements the
plot method, you can
visualize the environment behavior during training and simulation. If you call
plot(env) before training or simulation, where
is your environment object, then the visualization updates during training to allow you to
visualize the progress of each episode or simulation.
Environment visualization is not supported when training or simulating your agent using parallel computing.
For custom environments, you must implement your own
For more information on creating custom environments with plot functions, see Create Custom MATLAB Environment from Template.