Deep Reinforcement Learning Lecture
Nowadays, Artificial Intelligence, notably Advanced Machine Learning (ML) drives scientific and economic growth worldwide. They are essentially massive ‘learning by experience/examples’ systems. However, as our tasks and the world change, such systems should adapt to new domains/tasks and continue learning. Knowledge should be transferred from one DNN systems to other ones. Distributed DNN training should be performed though Federated Learning, e.g., for privacy protection. New Learning modes should be explored, by reward maximation, as it is done in Deep Reinforcement Learning and Imitation Learning.

This lecture overviews Deep Reinforcement Learning that has many applications in, e.g., Game playing agents, Self-driving vehicles, Robotics (Robot cleaners) and Stock exchange agents. It covers the following topics in detail: Finite Markov Decision Processes. Elements of RL (actions, states, Policy, Reward, Value function, Q-function). RL algorithms for finding the optimal policy: Dynamic Programming, Monte Carlo, Temporal-difference learning, SARSA, Q-learning. Deep RL algorithms, DQN and its extensions, Rainbow. Policy Gradient methods. Actor Critic Methods. Imitation Learning. A Maze example is also presented.