MATLABEnvironments

Model reinforcement learning environment dynamics using MATLAB^®

In a reinforcement learning scenario, the environment models the dynamics with which the agent interacts. The environment:

Receives actions from the agent
Outputs observations resulting from the dynamic behavior of the environment model
Generates a reward measuring how well the action contributes to achieving the task

You can create predefined and custom environments in MATLAB. For more information, seeCreate MATLAB Reinforcement Learning Environments.

Functions

`rlPredefinedEnv`	Create a predefined reinforcement learning environment
`rlFunctionEnv`	Specify custom reinforcement learning environment dynamics using functions
`rlCreateEnvTemplate`	Create custom reinforcement learning environment template
`rlMDPEnv`	Create Markov decision process environment for reinforcement learning
`createMDP`	Create Markov decision process model
`createGridWorld`	Create a two-dimensional grid world for reinforcement learning
`validateEnvironment`	Validate custom reinforcement learning environment

`generateRewardFunction`	Generate a reward function from control specifications to train a reinforcement learning agent
`exteriorPenalty`	Exterior penalty value for a point with respect to a bounded region
`hyperbolicPenalty`	Hyperbolic penalty value for a point with respect to a bounded region
`barrierPenalty`	Logarithmic barrier penalty value for a point with respect to a bounded region

`rlFiniteSetSpec`	Create discrete action or observation data specifications for reinforcement learning environments
`rlNumericSpec`	Create continuous action or observation data specifications for reinforcement learning environments
`getActionInfo`	Obtain action data specifications from reinforcement learning environment or agent
`getObservationInfo`	Obtain observation data specifications from reinforcement learning environment or agent

Model environment dynamics using a MATLAB object that interacts with the agent, generating rewards and observations in response to agent actions.

Import a custom MATLAB environment or create a predefined MATLAB environment.

Create a reward signal that measures how successful the agent is at achieving its goal.

Load predefined MATLAB control system environments.

You can train agents in predefined MATLAB grid world environments for which the actions, observations, and rewards are already defined.

You can create custom MATLAB grid world environments by defining your own size, rewards and obstacles.

Create a reinforcement learning environment by supplying custom dynamic functions.

You can define a custom reinforcement learning environment by creating and modifying a template environment object.