McKeon Richard. Neural Networks for Electronics Hobbyists

Apress, 2018. — 139 p. — ISBN-13 9781484235065.Learn how to implement and build a neural network with this non-technical, project-based book as your guide. As you work through the chapters, you'll build an electronics project, providing a hands-on experience in training a network.There are no prerequisites here and you won't see a single line of computer code in this book. Instead, it takes a hardware approach using very simple electronic components. You'll start off with an interesting non-technical introduction to neural networks, and then construct an electronics project. The project isn't complicated, but it illustrates how back propagation can be used to adjust connection strengths or "weights" and train a network.By the end of this book, you'll be able to take what you've learned and apply it to your own projects. If you like to tinker around with components and build circuits on a breadboard, Neural Networks for Electronics Hobbyists is the book for you.Biological Neural Networks
Biological Computing: The Neuron
What Did You Do to Me?
Wetware, Software, and Hardware
Applications
Just Around the Corner
Implementing Neural Networks
Architecture?
A Variety of Models
Our Sample Network
Training the NetworkElectronic Components
What Is XOR?
The Protoboard
The Power Supply
Inputs
What Is a Voltage Divider?
Adjusting Connection Weights
Summing Voltages
Op Amp Comparator
Putting It All Together
Parts ListBuilding the Network
Do We Need a Neural Network?
The Power Supply
The Input Layer
The Hidden Layer
The Output Layer
Testing the circuitTraining with Back Propagation
The Back Propagation Algorithm
Training Cycles
Convergence
Attractors and Trends
ImplementationTraining on Other Functions
The OR Function
The AND Function
The General Purpose MachineWhere Do We Go from Here?
Varying the Learning Rate
Crazy Starting Values
Apply the Back Propagation Rule Differently
Feature Extraction
Determining the Range of Values
Training on Different Logic Functions
Try Using a Different Model
Build a Neural Network to Do Other Things
PostscriptAppendix A: Neural Network Software, Simbrain
Appendix B: Resources
Neural Network Books
Chaos and Dynamic Systems