Today, we are launching Machine Learning 3: Reinforcement Learning! This is the third part of a Machine Learning series created with Georgia Tech as part of their Online Masters of Computer Science degree program.
We’re proud to bring this course and others created for the degree program on the Udacity site in the Georgia Tech Masters in CS track. You can take these courses (without applying for the degree program!) and learn by doing real world projects.
In the first two mini courses of the series, you’ll learn Supervised Learning and Unsupervised Learning. Now you can follow the fun-loving instructors, Professors Charles Isbell and Michael Littman, as they examine reinforcement learning concepts and techniques.
What is reinforcement learning? Reinforcement learning is about mapping situations to actions, and includes techniques like Game Theory and the Markov Decision Processes.
You may already use reinforcement learning when playing games like chess or poker! For instance, when a chess player makes a move, she is using reinforcement learning. Her choice is determined by planning (anticipating possible reactions) and by immediate, intuitive judgements of the desirability of particular positions and moves. Similarly, in Poker, a player balances risk and reward while deciding what decision to make.
In the final project, you’ll bring back the 90’s and design a Pacman agent capable of eating all the food without getting eaten by monsters.
Ready to get started? We hope you like the course. Happy (Machine) Learning!
