TensorFlow Beginner: Neural Networks

Name: TensorFlow Beginner: Neural Networks
Author: Josh Turner

Learn the foundational concepts of neural networks and explore the applications of TensorFlow in building them

Preview the first lesson free — get full access to all 3 lessons.

Course: On-Demand

Hello, my name is Josh Turner and these lessons you'll learn about different types of models. Let's get started. So the first one that we want to discuss are supervised models. Supervised models are models in which any time that you have labeled data for your training or output data. So let's say that I have a binary model where I'm trying to predict two different things. Maybe it's someone is sick or someone is not sick. In that case, I have labels for sick and not sick in my data set. And that would be an example of a supervised model. Supervised models come in many types. Let's just discuss a few of the types of supervised models. Any type of regression model can be considered as a supervised model. For example, artificial neural networks, those are supervised learning models. Time series where you're trying to predict some kind of output data, for example, sales or prevalence of disease, and decision trees are all examples of supervised models. More examples of supervised models further break down the types of machine learning methods. So for example, we have regression and classification models. Those are both types of supervised models, convolutional neural networks, support vector machines, and again, you see decision trees as both regression and classification. Those are all types of supervised learning models. So now we know and understand what it means to be a supervised model. It's any kind of model where your data has a label and you're trying to predict something about that label. The other type of model that we can look at is unsupervised learning models. There are many different types of unsupervised learning models as well. I'll tell you many of the names of these models. We won't go over them in depth, but at least you'll know the names of the models for future reference. Unsupervised models don't come with labels, so we may not be trying to understand or predict different types of labels. Some different examples of unsupervised models include the following. The K-means model. Clustering models are good examples of unsupervised models. PCA or principal component analysis. A model called DB scan, which is a density clustering type of model. Outlier methods, which you may have seen before, those are also unsupervised learning models, and really different types of density-based estimation can be seen as unsupervised learning models. Okay, so those are some different types of unsupervised learning models. Next let's talk about two different classes of models. The first one that we'll talk about are parametric models. Parametric models are any kind of models where you have some kind of parameters to tune. So for example, you might be tuning the parameters of weights and biases of a neural network. That would be an example of a parametric model. Other examples include convolutional neural networks and basic linear regression. In contrast, we have a type of model called non-parametric models. Non-parametric models are the kind of models you create when you use basic descriptive statistics, and a good example of that is Bayesian inference. Thanks for watching. Stay tuned for the next lesson where you'll learn about artificial neural networks.

Beginner Provider : Josh Turner 3 Lessons · 19m in Arabic, German, English, Spanish, French, Portuguese, Chinese

Course Description

Neural networks have become an integral part of machine learning and can be used in a variety of applications, from image processing to natural language processing. In these lessons, you will learn about the foundational concepts of neural networks, including Artificial Neural Networks (ANNs) and the Least Squares method. You will explore how to build an ANN using TensorFlow and how to use the Least Squares method to optimize its performance.

Moving forward, you will delve into Convolutional Neural Networks (CNNs), a type of neural network commonly used in image processing. You will learn about the different layers used in a CNN, such as the Convolutional layer, Pooling layer, and Flatten layer, and their primary uses in image processing. You will also discover the different applications of CNNs, such as image recognition and object detection.

Overall, you will gain a foundational understanding of the different concepts and techniques used in building neural networks and the skills necessary to apply them in your projects.

What You'll Learn

Understand the foundational concepts of neural networks, including Artificial Neural Networks (ANNs) and the Least Squares method
Build an Artificial Neural Network (ANN) using TensorFlow
Apply the Least Squares method to optimize the performance of an ANN
Explore Convolutional Neural Networks (CNNs) and the layers used in them, including the Convolutional, Pooling, and Flatten layers
Identify applications of CNNs such as image recognition and object detection

Key Takeaways

Neural networks are an integral part of machine learning and can be used in applications ranging from image processing to natural language processing.
Artificial Neural Networks (ANNs) can be built using TensorFlow, and the Least Squares method can be used to optimize their performance.
Convolutional Neural Networks (CNNs) are a type of neural network commonly used in image processing.
A CNN uses different layers, including the Convolutional layer, Pooling layer, and Flatten layer, each with a primary use in image processing.
CNNs have applications such as image recognition and object detection.

Frequently Asked Questions

What will I learn in this course?

You will learn the foundational concepts of neural networks, including Artificial Neural Networks (ANNs) and the Least Squares method, how to build an ANN using TensorFlow, how to optimize its performance with the Least Squares method, and the foundations of Convolutional Neural Networks (CNNs) used in image processing.

Is this course suitable for beginners?

Yes. The course provides a foundational understanding of the different concepts and techniques used in building neural networks and the skills necessary to apply them in your projects.

What are Convolutional Neural Networks (CNNs) covered in this course?

CNNs are a type of neural network commonly used in image processing. The course covers the different layers used in a CNN, such as the Convolutional layer, Pooling layer, and Flatten layer, and applications such as image recognition and object detection.

What topics do the lessons cover?

The lessons cover Types of Models, Artificial Neural Networks, and Convolutional Neural Networks.

What skills will I gain from this course?

You will build skills in Artificial Neural Networks, Deep Learning, Deep Learning Methods, Imagenet, Neural Engineering, and TensorFlow.