Neural networks are powerful predictive tools that can be used for almost any machine learning problem with very good results. If you want to break into deep learning and artificial intelligence, learning neural networks is the first crucial step.
This is why I’m inviting you to an exciting journey through the world of complex, state-of-the-art neural networks. In this course you will develop a strong understanding of the most utilized neural networks, suitable for both classification and regression problems.
The mathematics behind neural networks is particularly complex, but you don’t need to be a mathematician to take this course and fully benefit from it. We will not dive into complicated maths – our emphasis here is on practice. You will learn how to operate neural networks using the R program, how to build and train models and how to make predictions on new data.
All the procedures are explained live, on real life data sets. So you will advance fast and be able to apply your knowledge immediately.
This course contains four comprehensive sections.
1. Multilayer Perceptrons – Beyond the Basics
Learn to use multilayer perceptrons to make predictions for both categorical and continuous variables. Moreover, learn how to test your models accuracy using the k-fold cross-validation technique and how improve predictions by manipulating various parameters of the network.
2. Probabilistic Neural Networks
These networks are primarily used for classification problems, so we’ll learn how to predict a bank’s customers default with their help. Next, we’ll see how to look for optimal values of the smoothing parameter in order to make more accurate predictions.
3. Generalized Regression Neural Networks
If you have to solve a regression problem (where your response variable is numeric), these networks can be very effective. We’ll show how to predict a car value based on its technical characteristics and how to improve the prediction by controlling the smoothing parameter of our model. The k-fold cross-validation techniques will also be employed to identify better models.
4. Recurrent Neural Networks
These networks are useful for many prediction problems, but they are particularly valuable for time series modelling and forecasting. In this course we focus on two types of recurrent neural networks: Elman and Jordan. We are going to use them to predict future air temperatures based on historical data. Making truthful predictions on time series is generally very tough, but we will do our best to build good quality models and get satisfactory values for the prediction accuracy metrics.
For each type of network, the presentation is structured as follows:
-
a short, easy to understand theoretical introduction (without complex mathematics)
-
how to train the network in R
-
how to test the network to make sure that it does a good prediction job on independent data sets.
For every neural network, a number of practical exercises are proposed. By doing these exercises you’ll actually apply in practice what you have learned.
This course is your opportunity to become a neural network expert in a few days only (literally). With my video lectures, you will find it very easy to master these major neural network and build them in R. Everything is shown live, step by step, so you can replicate any procedure at any time you need it.
So click the “Enroll” button to get instant access to your course. It will surely get you some new, valuable skills. And, who knows, it could greatly enhance your future career.
See you inside!
Multilayer Perceptron - Beyond the Basics
Probabilistic Neural Networks
-
2What Are Multilayer Perceptrons?
A quick view in the world of (maybe) the most popular neural networks - multilayer perceptrons. What they are and how their structure looks like.
-
3How Multilayer Perceptrons Work?
Let's learn more about summation and activation, the more important processes that take place in a neural network.
-
4How Does a Multilayer Perceptron Learn?
The learning process of a multilayer perceptron, in layman's terms.
-
5Prediction Accuracy Metrics
How to measure the predictive performance of MLPs - a few simple indicators.
-
6ROC Curve
Here's another powerful tool for assessing the prediction performance of a neural network when the response variable is categorical: the ROC curve. Let's see what it is and how to interpret it.
-
7Using MLPs With Categorical Response Variables: Building the Network
How to set the parameters of a multilayer perceptron when the response variable is qualitative. And how to prepare your data properly.
-
8Using MLPs With Categorical Response Variables: Making Predictions
How to use our MLP to make predictions in the test set.
-
9Using MLPs With Categorical Response Variables: ROC Curve
Learn the code for creating ROC curves in R and computing the area under curve (it's simple).
-
10Using MLPs With Categorical Response Variables: Playing With the Hidden Nodes
Let's try to improve the prediction accuracy of our network by changing the number of hidden nodes in the hidden layer.
-
11Using MLPs With Categorical Response Variables: K-Fold Validation
Finally, let's validate our predictive model using the k-fold validation technique. Learn a few commands in the caret package that make k-fold validation very easy.
-
12Using MLPs With Continuous Response Variables: Building the Network
How to set the parameters of MLP when the response variable is numeric.
-
13Using MLPs With Continuous Response Variables: Making Predictions
How to make predictions and assess the predictive accuracy of our MLP in the test set.
-
14Using MLPs With Continuous Response Variables: Manipulating the Hidden Nodes
Let's see if we can improve the predictive performance of our network by modifying the number of nodes in the hidden layer.
-
15Using MLPs With Continuous Response Variables: K-Fold Validation
Validating our MLP using the k-fold technique.
Generalized Regression Neural Networks
-
16What Are Probablistic Neural Networks?
A short theoretical introduction: the structure of a probabilistic neural network and how it works.
-
17Data Preparation
How to prepare your data properly so you can use it as an input for a PNN.
-
18Building the Network
How to create a probabilistic neural network in R, step by step.
-
19Making Predictions
How to make predictions in the training set using our new PNN.
-
20Finding the Optimal Sigma
Improve the prediction accuracy of our PNN by changing the value of the smoothing parameter sigma.
-
21Validating Our Model
After identifying an optimal sigma, we validate our PNN using the k-fold cross-validation technique.
Recurrent Neural Networks
-
22What Are Generalized Regression Neural Networks?
A short theoretical introduction: the structure of a generalized regresssion neural network and how it works.
-
23Data Preparation
How to arrange your data so it can be used for GRNNs.
-
24Building the Network
All the commands you need to build and train a GRNN in the R program.
-
25Making Predictions
Let's evaluate the prediction accuracy of our new GRNN in the training set.
-
26Finding the Optimal Sigma
Identify the smoothing parameter value that maximizes the prediction accuracy of our network.
Practice
-
27What Are Recurrent Neural Networks?
A short theoretical introduction: the structure of a recurrent neural network and how it works.
-
28Measuring the Predictive Performance
In this lecture we introduce two specific metric for measuring the predictive performance in time series: mean absolute standard error and mean absolute percentage error.
-
29Elman Networks: Data Preparation
How to prepare your data for creating an Elman recurring neural network. We are going to use a package called quantmod to create lagged variables.
-
30Elman Networks: Building the Model
How to build and train an Elman network in R.
-
31Elman Networks: Making Predictions
How to do times series forecasting with our new Elman network, using specific indicators.
-
32Elman Networks: Adding More Predictors
We'll try to improve the prediction performance of our Elman network by adding more lagged variables as predictors.
-
33Elman Networks: Making Predictions With Our New Model
Assessing the predictive accuracy in the training set after adding more variables to the model.
-
34Jordan Networks: Data Preparation
Prepare your data so you can use it as an input in a Jordan recurrent neural network.
-
35Jordan Networks: Building the Model
Time to build and train our Jordan recurrent network.
-
36Jordan Networks: Making Predictions
Let's perform time series forecasting with our Jordan network.