Reinforcement Learning Designer

Design, train, and simulate reinforcement learning agents

Description

The Reinforcement Learning Designer app lets you design, train, and simulate agents for existing environments.

Using this app, you can:

• Import an existing environment from the MATLAB^® workspace or create a predefined environment.

• Automatically create or import an agent for your environment (DQN, DDPG, TD3, SAC, and PPO agents are supported).

• Train and simulate the agent against the environment.

• Analyze simulation results and refine your agent parameters.

• Export the final agent to the MATLAB workspace for further use and deployment.

Limitations

The following features are not supported in the Reinforcement Learning Designer app.

• Multiagent systems

• Q, SARSA, PG, AC, and SAC agents

• Custom agents

• Agents relying on table or custom basis function representations

If your application requires any of these features then design, train, and simulate your agent at the command line.

Open the Reinforcement Learning Designer App

• MATLAB Toolstrip: On the Apps tab, under Machine Learning and Deep Learning, click the app icon.

• MATLAB command prompt: Enter reinforcementLearningDesigner.

Examples

• Design and Train Agent Using Reinforcement Learning Designer
• Train Reinforcement Learning Agents

Parameters

Use reinforcementLearningDesigner(SessionFile) to open the Reinforcement Learning Designer app with the design session saved in SessionFile.

Programmatic Use

reinforcementLearningDesigner opens the Reinforcement Learning Designer app. You can then import an environment and start the design process, or open a saved design session.

Version History

Introduced in R2021a