Have you wondered how Google DeepMind was able to get expert level results in Atari in a short amount of time with only the screen as input? This talk will go into details about Reinforcement Learning and the Q-Learning algorithm that made it possible. Using just a few lines of TensorFlow and Keras I will show how to implement these various iterations of the algorithm.
The meeting will also cover some of the newer features coming our of Google. Including TensorFlow v2 and the RL_Agent library.
Speaker: Evan Hennis is a recent graduate of Georgia Tech with a Master's degree in Computer Science with a specialization in Machine Learning. Recently, he became a Google Developer Expert (GDE) in Machine Learning because of his contributions to the community. These contributions were keeping up a blog (https://eckronsoftware.wordpress.com), a YouTube channel (Evan Hennis), speaking at conferences, and contributing to the TensorFlow open source project.
Topics: TensorFlow, Keras, Q-Learner, Double Q-Learner, Deep Q-Network, DDQN, RL_Agent, Deepmind and more
There will be pizza and networking to kick off the event