State of the art AI Methods with TensorFlow: Transfer Learning, RL and GANs
State of the art AI Methods with TensorFlow: Transfer Learning, RL and GANs

Abstract: 

Although supervised learning has dominated industry machine learning implementations, unsupervised and semi-supervised methods have started to be practically applied to real world problems (outside of playing video games). Generative Adversarial Networks (GANs) are being utilized to augment data and generate dialogue, and Reinforcement Learning (RL) is helping people plan marketing campaigns and control robots.

In this training, you will develop a theoretical understanding of these and other related state-of-the-art AI methods along with the hands-on skills needed to train and utilize them. You will implement a variety of models in TensorFlow for tasks including object recognition, image generation and robotics.

Session Outline
Module 1: Transfer Learning
There are so many pre-trained models out there from brilliant research groups, and this means that you don't always have to start your AI development from scratch. This module will teach you how to fine-tune AI models from a pre-trained parent model for faster training times and better performance.

Module 2: Reinforcement Learning
Learn about sequential decision making in an environment that responds to your actions. In this module, you will learn about Reinforcement Learning and Deep Q-Learning, and you will implement these methods using TensorFlow.

Module 3: GANs
Sometimes you need to generate new samples in a certain style that are indistinguishable from human-generated samples. In this module, we will explore Generative Adversarial Networks (GANs), and you will implement one of these networks to generate human-like handwriting.

Background Knowledge
Python programming, familiarity with Jupyter notebooks, foundational math knowledge (some exposure to things like exponentials and matrix arithmetic), experience with the basic machine learning workflow (pre-processing, training, testing, inference).

Bio: 

Daniel Whitenack (aka Data Dan) is a PhD trained data scientist who has been developing artificial intelligence applications in the real world for over 10 years. He knows how to see beyond the hype of AI and machine learning to build systems that create business value, and he has taught these skills to 1000’s of developers, data scientists, and engineers all around the world. Now with the AI Classroom event, Data Dan is bringing this knowledge to an live, online learning environment so that you can level up your career from anywhere!