A reinforcement learning agent receives observations and a reward from the environment. Using its policy, the agent selects an action based on the observations and reward, and sends the action to the environment. During training, the agent continuously updates the policy parameters based on the action, observations, and reward. Doing so, allows the agent to learn the optimal policy for the given environment and reward signal.
Reinforcement Learning Toolbox™ software provides reinforcement learning agents that use several common algorithms, such as SARSA, DQN, DDPG, and A2C. You can also implement other agent algorithms by creating your own custom agents.
For more information, see Reinforcement Learning Agents. For more information on defining policy representations, see Create Policy and Value Function Representations.
You can create an agent using one of several standard reinforcement learning algorithms or define your own custom agent.
Create Q-learning agents for reinforcement learning.
Create SARSA agents for reinforcement learning.
Create DQN agents for reinforcement learning.
Create PG agents for reinforcement learning.
Deep Deterministic Policy Gradient Agents
Create DDPG agents for reinforcement learning.
Twin-Delayed Deep Deterministic Policy Gradient Agents
Create DDPG agents for reinforcement learning.
Create AC agents for reinforcement learning.
Proximal Policy Optimization Agents
Create PPO agents for reinforcement learning.
Create SAC agents for reinforcement learning.
Create agents that use custom reinforcement learning algorithms.