Training and Validation
To learn an optimal policy, a reinforcement learning agent interacts with the environment through a repeated trial-and-error process. During training, the agent tunes the parameters of its policy representation to maximize the long-term reward. Reinforcement Learning Toolbox™ software provides functions for training agents and validating the training results through simulation. For more information, see Train Reinforcement Learning Agents.
|Reinforcement Learning Designer||Design, train, and simulate reinforcement learning agents|
|Train reinforcement learning agents within a specified environment|
|Options for training reinforcement learning agents|
|Simulate trained reinforcement learning agents within specified environment|
|Options for simulating a reinforcement learning agent within an environment|
|Plot training information from a previous training session|
|RL Agent||Reinforcement learning agent|
Training and Simulation Basics
Find the optimal policy by training your agent within a specified environment.
Train Q-learning and SARSA agents to solve a grid world in MATLAB®.
Train a reinforcement learning agent in a generic Markov decision process environment.
Train a controller using reinforcement learning with a plant modeled in Simulink® as the training environment.
Using the Reinforcement Learning Designer App
Design and train a DQN agent for a cart-pole system using the Reinforcement Learning Designer app.
Interactively specify options for simulating reinforcement learning agents.
Interactively specify options for training reinforcement learning agents.
Using Multiple Processes and GPUs
Accelerate agent training by running simulations in parallel on multiple cores, GPUs, clusters or cloud resources.
Train actor-critic agent using asynchronous parallel computing.
Train a reinforcement learning agent for an automated driving application using parallel computing.
Train Agents in MATLAB Environments
Train a deep deterministic policy gradient agent to control a second-order dynamic system modeled in MATLAB.
Train a policy gradient with a baseline to control a double integrator system modeled in MATLAB.
Train a deep Q-learning network agent to balance a cart-pole system modeled in MATLAB.
Train a policy gradient agent to balance a cart-pole system modeled in MATLAB.
Train an actor-critic agent to balance a cart-pole system modeled in MATLAB.
Train a reinforcement learning agent using an image-based observation signal.
Create a reinforcement learning agent using the Deep Network Designer app from the Deep Learning Toolbox™.
Train Agents in Simulink Environments
Train a Deep Q-network agent to balance a pendulum modeled in Simulink.
Train a deep deterministic policy gradient agent to balance a pendulum modeled in Simulink.
Train a reinforcement learning agent to balance a pendulum Simulink model that contains observations in a bus signal.
Train a deep deterministic policy gradient agent to swing up and balance a cart-pole system modeled in Simscape™ Multibody™.
Train two PPO agents to collaboratively move an object.
Train three PPO agents to explore a grid-world environment in a collaborative-competitive manner.
Train a DQN and a DDPG agent to collaboratively perform adaptive cruise control and lane keeping assist to follow a path.
Generate rewards from control specifications
Generate a reward function from an MPC controller applied to a servomotor.
Generate a reward function from an model verification block applied to a water tank system.
Train a deep neural network to imitate the behavior of a model predictive controller within a lane keeping assist system.
Train a deep neural network to imitate the behavior of a nonlinear model predictive controller for a flying robot.
Train a reinforcement learning agent using an actor network that has been previously trained using supervised learning.
Custom Agents and Training Algorithms
Train a custom LQR agent.
Train a reinforcement learning policy using your own custom training algorithm.
Create agent for custom reinforcement learning algorithm.