Tensorflow and Deep Learning - Without a PhD

Dive into a comprehensive conference talk on TensorFlow and deep learning, designed for beginners and experienced developers alike. Learn how to choose the right neural network for your problem and make it behave without requiring advanced mathematical knowledge. Explore fully connected and convolutional neural networks, regularization techniques, recurrent neural networks, natural language analysis, word embeddings, transfer learning, and image analysis and generation. Gain practical insights through numerous examples and discover how to implement deep learning solutions using Python and TensorFlow. Follow along as the speaker covers essential topics such as matrix multiplication, training systems, placeholders, variables, gradient descent, layers, and regularization techniques. By the end of this talk, acquire the skills to transition from machine learning research to practical software engineering applications.

Introduction
Outline
Neural network
Matrix multiply
Recap
Training the system
Training results
Placeholders and variables
Model
Gradient Descent
Weights
Training code recap
Layers
The Relu
The Problem
Results
Accuracy
Regularization
Zero