Session

Oleksandr Maksymets
13 Oct 2018
10:00 - 11:40
Workshop

A Glance at Reinforcement Learning

Latest achievements of AI beating human performance on Go and Dota 2 games were powered by Reinforcement Learning (RL). We will touch the basics of RL such as RL problem definition, different value functions, policy. The brief overview of main RL algorithms including Q-learning, Actor-Critic Algorithm. We will dive deeper in implementation one of the algorithms and run it on OpenAI gym environment. Finish with a summary of practical and business applications of RL.

Minimal requirements to the listener: knowledge of Python, basics of probability theory and machine learning.
Repository: https://github.com/mathfac/aiukraine_rl_workshop

Presentation