T81 558:Applications of Deep Neural Networks

Washington University in St. Louis

Instructor: Jeff Heaton

The content of this course changes as technology evolves, to keep up to date with changes follow me on GitHub.

Fall 2019, Mondays, Section 1: 2:30P-5:20P, Section 2: 6-9:00P, Online and in class room: TBA

Course Description

Deep learning is a group of exciting new technologies for neural networks. Through a combination of advanced training techniques and neural network architectural components, it is now possible to create neural networks that can handle tabular data, images, text, and audio as both input and output. Deep learning allows a neural network to learn hierarchies of information in a way that is like the function of the human brain. This course will introduce the student to classic neural network structures, Convolution Neural Networks (CNN), Long Short-Term Memory (LSTM), Gated Recurrent Neural Networks (GRU), General Adversarial Networks (GAN) and reinforcement learning. Application of these architectures to computer vision, time series, security, natural language processing (NLP), and data generation will be covered. High Performance Computing (HPC) aspects will demonstrate how deep learning can be leveraged both on graphical processing units (GPUs), as well as grids. Focus is primarily upon the application of deep learning to problems, with some introduction to mathematical foundations. Students will use the Python programming language to implement deep learning using Google TensorFlow and Keras. It is not necessary to know Python prior to this course; however, familiarity of at least one programming language is assumed. This course will be delivered in a hybrid format that includes both classroom and online instruction.

Objectives

1. Explain how neural networks (deep and otherwise) compare to other machine learning models.
2. Determine when a deep neural network would be a good choice for a particular problem.
3. Demonstrate your understanding of the material through a final project uploaded to GitHub.

Syllabus

This syllabus presents the expected class schedule, due dates, and reading assignments. Download current syllabus.

Module	Content
Module 1 Meet on 08/26/2019	Part 1.1: Course Overview Part 1.2: Introduction to Python Part 1.3: Python Lists, Dictionaries, Sets & JSON Part 1.4: File Handling Part 1.5: Functions, Lambdas, and Map/ReducePython Preliminaries We will meet on campus this week! (first meeting)
Module 2 Week of 09/09/2019	Part 2.1: Introduction to Pandas for Deep Learning Part 2.2: Encoding Categorical Values in Pandas Part 2.3: Grouping, Sorting, and Shuffling Part 2.4: Using Apply and Map in Pandas Part 2.5: Feature Engineering in Padas Module 1 Assignment Due: 09/10/2019
Module 3 Week of 09/16/2019	Part 3.1: Deep Learning and Neural Network Introduction Part 3.2: Introduction to Tensorflow & Keras Part 3.3: Saving and Loading a Keras Neural Network Part 3.4: Early Stopping in Keras to Prevent Overfitting Part 3.5: Extracting Keras Weights and Manual Neural Network Calculation Module 2: Assignment due: 09/17/2019
Module 4 Week of 09/23/2019	Part 4.1: Encoding a Feature Vector for Keras Deep Learning Part 4.2: Keras Multiclass Classification for Deep Neural Networks with ROC and AUC Part 4.3: Keras Regression for Deep Neural Networks with RMSE Part 4.4: Backpropagation, Nesterov Momentum, and ADAM Training Part 4.5: Neural Network RMSE and Log Loss Error Calculation from Scratch Module 3 Assignment due: 09/24/2019
Module 5 Meet on 09/30/2019	Part 5.1: Introduction to Regularization: Ridge and Lasso Part 5.2: Using K-Fold Cross Validation with Keras Part 5.3: Using L1 and L2 Regularization with Keras to Decrease Overfitting Part 5.4: Drop Out for Keras to Decrease Overfitting Part 5.5: Bootstrapping and Benchmarking Hyperparameters Module 4 Assignment due: 10/01/2019 We will meet on campus this week! (2nd Meeting)
Module 6 Week of 10/07/2019	Part 6.1: Image Processing in Python Part 6.2: Keras Neural Networks for MINST and Fashion MINST Part 6.3: Implementing a ResNet in Keras Part 6.4: Computer Vision with OpenCV Part 6.5: Recognizing Multiple Images with Darknet Module 5 Assignment due: 10/08/2019
Module 7 Week of 10/14/2019	Part 7.1: Introduction to GANS for Image and Data Generation Part 7.2: Implementing a GAN in Keras Part 7.3: Face Generation with StyleGAN and Python Part 7.4: GANS for Semi-Supervised Learning in Keras Part 7.5: An Overview of GAN Research Module 6 Assignment due: 10/15/2019
Module 8 Meet on 10/21/2019	Part 8.1: Introduction to Kaggle Part 8.2: Building Ensembles with Scikit-Learn and Keras Part 8.3: How Should you Architect Your Keras Neural Network: Hyperparameters Part 8.4: Bayesian Hyperparameter Optimization for Keras Part 8.5: Current Semester's Kaggle Module 7 Assignment due: 10/22/2019 We will meet on campus this week! (3rd Meeting)
Module 9 Week of 10/28/2019	Part 9.1: Introduction to Keras Transfer Learning Part 9.2: Popular Pretrained Neural Networks for Keras. Part 9.3: Transfer Learning for Computer Vision and Keras Part 9.4: Transfer Learning for Languages and Keras Part 9.5: Transfer Learning for Keras Feature Engineering Module 8 Assignment due: 10/29/2019
Module 10 Week of 11/04/2019	Part 10.1: Time Series Data Encoding for Deep Learning, TensorFlow and Keras Part 10.2: Programming LSTM with Keras and TensorFlow Part 10.3: Image Captioning with Keras and TensorFlow Part 10.4: Temporal CNN in Keras and TensorFlow Part 10.5: Predicting the Stock Market with Keras and TensorFlow Module 9 Assignment due: 11/05/2019
Module 11 Week of 11/11/2019	Part 11.1: Getting Started with Spacy in Python Part 11.2: Word2Vec and Text Classification Part 11.3: Natural Language Processing with Spacy and Keras Part 11.4: What are Embedding Layers in Keras Part 11.5: Learning English from Scratch with Keras and TensorFlow Module 10 Assignment due: 11/12/2019
Module 12 Meet on 11/18/2019	Kaggle Assignment due: 11/17/2019 (approx 4-6PM, due to Kaggle GMT timezone) We will meet on campus this week! (4th Meeting) Part 12.1: Introduction to the OpenAI Gym Part 12.2: Introduction to Q-Learning for Keras Part 12.3: Keras Q-Learning in the OpenAI Gym Part 12.4: Atari Games with Keras Neural Networks Part 12.5: How Alpha Zero used Reinforcement Learning to Master Chess
Module 13 Week of 11/25/2019	Part 13.1: Deploying a Model to AWS Part 13.2: Flask and Deep Learning Web Services Part 13.3: AI at the Edge: Using Keras on a Mobile Device Part 13.4: When to Retrain Your Neural Network Part 13.5: Using a Keras Deep Neural Network with a Web Application
Module 14 Week of 12/02/2019	Part 14.1: What is AutoML Part 14.2: Using Denoising AutoEncoders in Keras Part 14.3: Anomaly Detection in Keras Part 14.4: Training an Intrusion Detection System with KDD99 Part 14.5: The Deep Learning Technologies I am Excited About Final Project due 12/09/2019

Datasets

• Iris - Classify between 3 iris species.
• Auto MPG - Regression to determine MPG.
• WC Breast Cancer - Binary classification: malignant or benign.
• toy1 - The toy1 dataset, regression for weights of geometric solids.

Note: Other datasets may be added as the class progresses.

Final Project

For the final project you can choose a security project or choose your own dataset to create and fit a neural network. For more information:

• Security Project - See Canvas for more information.
• Independent Project - Choose your own dataset or one of my suggestions.

Other Information

• Helpful Functions - Helpful Python functions for this class.
• KDD99 Example
• Care and Feeding of Python - Some useful commands for a local Python install. Not needed if you are using Data Scientist Workbench.