This masterclass discussed the implementation of autonomous agents for optimal control using AI, ML and DRL techniques. To this end, we looked at how to model the challenge and the environment as a Markov Decision Process (MDP) and its resolution using the main RL algorithms such as Q-learning and/or its derivatives.
We also saw how the combined use of reinforcement learning techniques and approximation functions gleaned, for example, from deep learning enables us to respond to problems in complex environments.
Addressed to:
People with basic knowledge of artificial intelligence, data analysts or data scientists with an interest in understanding how reinforcement learning techniques work.
The approach will be theoretical.
