Deep Learning with TensorFlow

Explore the fundamentals of TensorFlow in this comprehensive conference talk. Dive into an interactive discussion covering basic elements of the framework and their composition for creating deep learning machine learning models. Gain a strong foundation in both creating and consuming deep learning models in applications. Begin with an introduction to AI and machine learning concepts, including accuracy vs recall and bias. Progress through programming concepts, linear discriminators, and model outputs. Understand key mathematical concepts like mean squared error, gradient descent, and mathematical induction. Delve into TensorFlow specifics, including TensorFlow flow, accuracy measurement, and training models. Learn about optimization graphs, freezing models, and inference. Explore advanced topics such as linear models, activation functions, derivatives, backpropagation, and multilayer perceptrons. Conclude with an overview of the Atom Optimizer. No prior exposure to TensorFlow is required, though a basic understanding of deep learning concepts is helpful.

Intro
AI and Machine Learning
Accuracy vs Recall
Programming
Bias
Linear Discriminator
Model
Output
H function
Mean squared error
Parabola example
Gradient Descent
Tensor Flow
TensorFlow Flow
Accuracy
Visual Studio
Training the model
Summary operation
Logs
Optimization Graph
Freezing a Model
Inference
Linear Model
Mathematical Induction
Activation Function
Derivatives
BackPropagation
Multilayer Perceptron
Atom Optimizer