Reinforcement learning (RL) is a branch of machine learning that achieves excellent results when applied to very complex problems. This approach to artificial intelligence is based on learning through the mechanism of rewards and punishments. There is a number of algorithms used in reinforcement learning, like Q-learning, Deep Q-networks and others. In this talk we will compare these algorithms and see when each of them should be used. Reinforcement learning is used in robotics, healthcare, resource management, gaming etc. An excellent example of applied reinforcement learning is AlphaGo, a first application that won a match against the world champion in Go, a strategy game previously considered out of reach for artificial intelligence. Join us and find out what are the advantages and disadvantages of reinforcement learning.