What you will learn in this course?
About this course
In this course, you’ll work with more complex environments, specifically provided by the OpenAI Gym: CartPole Mountain Car Atari games to train effective learning agents so you’ll need new techniques. It’s all about the application of neural networks to reinforcement learning and deep learning.
We’ve seen that reinforcement learning is an entirely different kind of machine learning than supervised and unsupervised learning.Supervised and unsupervised machine learning algorithms are for making predictions about data and analyzing, while reinforcement learning is about training an agent to interact with an environment and maximize its reward. Deep reinforcement learning and AI has a lot of potentials also carries huge risk.
One main principle of training reinforcement learning agents is that there are unintended consequences when training an AI. AI’s don’t think like humans, they come up with novel and non-intuitive solutions to reach their goals; often in ways that surprise domain experts.One of the great things about OpenAI is that they have a platform called the OpenAI Gym (heavy use of in this course). In order to train their reinforcement learning agents in the standard environments, it allows everyone and everywhere in this world.
Requirements-Basics of reinforcement learning, Dynamic Programming, TD Learning,Monte Carlo,MDPs; Knowledge of Calculus and probability(Undergraduate level); Experience in building models in machine learningPython and Numpy; Know how to build a feedforward, recurrent neural network using Theano and Tensorflow; convolutional.
Click the link or image below to access the course contents: