What Is Reinforcement Learning Toolbox?

Reinforcement Learning Toolbox™ provides MATLAB® functions and Simulink® blocks for training policies using reinforcement learning algorithms including DQN, A2C, and DDPG. The toolbox lets you implement controllers and decision-making systems for complex applications such as robotics, self-driving cars, and more.

You can represent policies and value functions using deep neural networks, polynomials, or lookup tables. Train policies by enabling reinforcement learning agents to interact with environments created in MATLAB or Simulink. Evaluate built-in and custom algorithms, experiment with hyperparameter settings, and monitor training progress. Accelerate training by parallelizing simulations and calculations on multicore CPUs, GPUs, computer clusters, and cloud resources (with Parallel Computing Toolbox™ and MATLAB Parallel Server™).

You can import existing policies from deep learning frameworks such as TensorFlow™ Keras and PyTorch through the ONNX™ model format (with Deep Learning Toolbox™). Generate optimized C, C++, and CUDA code to deploy trained policies on embedded platforms. The toolbox includes reference examples for using reinforcement learning to design controllers for robotics and automated driving applications.

Download ebook: Reinforcement Learning with MATLAB: Basics and Environment: https://bit.ly/2RS5Zrc