Course 806:
Introduction to Neural Networks and Deep Learning

(3 days)


Course Description

You do not need an extensive math background to understand neural network. In understandable steps, this course builds from a one node neural network to a multiple features, multiple output neural networks. All the steps are explained using working code to solve problems. After an understanding of how neural networks work and the parameters that control deep learning systems, Keras is introduced and used to simplify the building of deep learning neural networks. A convolutional deep learning neural network is built using Keras to show how deep learning is used in specialized neural networks. This course provides the necessary required background to understand ROI’s Time Series Analysis and Natural Language Processing courses.

Learning Objectives

In this course, attendees will:

  • Understand the context of neural networks and deep learning
  • Know how to use a neural network
  • Understand the data needs of deep learning
  • Have a working knowledge of neural networks and deep learning
  • Explore the parameters for neural networks


Familiarity with running programs from the command line is helpful, but not necessary. Understanding the concept of a function is helpful, but not necessary. NOTE: The course developers, having worked through the math, strongly believe using the models do not require understanding of the math used to implement the model.

Who Should Attend

Anyone who wants to understand Neural Network and Deep Learning.

Course Outline

Unit 1: The Simplest Possible Neural Network

  • What Is Machine Learning?
  • What Is a Neural Network?
  • Building the Simplest Neural Network in Simple Python
  • Multiple Input
  • Multiple Outputs
  • Use NumPy to Build Neural Networks

Unit 2: Updating Weights in Simplest Neural Network

  • Simple Error Analysis
  • Working with 1 Attributes
  • Small Steps
  • Extending Simplest Neural Network to Multiple Inputs
  • Extending to Multiple Outputs
  • Combining Multiple Input and Outputs

Unit 3: Extending to Complete Data Sets

  • Error vs. Cost
  • Extending Neural Network to Use Multiple Samples
  • Goodness of Fit Parameters

Unit 4: Understanding Back Propagation

  • Review of NumPy Arrays
  • Introduction to Stacked Arrays
  • Extending
  • Backpropagation
  • Coding Examples

Unit 5: Multiple Layers and Back Propagation

  • Introduction to Deep Learning
  • Forward Propagation
  • Back Propagation
  • Working Example

Unit 6: Parameters Affecting Deep Learning

  • Normalization
  • Data Size
  • Regularization
  • Weight Initialization
  • Working Though Coding Changes

Unit 7: Introduction of Keras

  • Why Keras?
  • Introduction to Linear Keras
  • Working Examples

Unit 8: Using DL for Vision – Convolution Neural Networks

  • The Problems of Pictures
  • A Solution
  • Implementing Solution in Keras
  • What You Really Need to Know If You Use Keras

Please Contact Your ROI Representative to Discuss Course Tailoring!