Build self-learning algorithms that optimize processes and problems
- Get an overview of reinforcement learning (RL) techniques in one day;
- Get familiar the latest Deep RL techniques as used by DeepMind.
This 1-day masterclass brings you up to speed in Deep Reinforcement Learning. Deep Reinforcement Learning is a new AI techniques that learns to optimize processes and problems. You will get an overview of reinforcement learning trends and techniques from both lectures and exercises.
The one day masterclass ‘Reinforcement Learning’ is intended for software and hardware engineers, game developers, robot engineers, application and process engineers, control engineers and system architects.
Prerequisites are mathematics skills and basic (python) programming skills.
After this 1-day masterclass the participants will have an understanding of reinforcement learning techniques. More specifically, they will:
- Understand the reinforcement learning framework;
- Understand solving the RL by dynamic programming;
- Understand solving the RL by Monte Carlo simulation;
- Understand solving the RL by Temporal Difference learning;
- Understand difference between tabular and function approximation approach;
- Understand Deep Reinforcement Learning, such as DQN, DDQN, BDQN;
- Understand replay memory techniques;
- Carry out reinforcement learning exercises, including training an algorithm to play Atari games.