Introducing artificial neural networks
The fundamental building block of an artificial neural network is the neuron. By itself, a single neuron is useless, but it can have strong predictive power when combined into a more complex network.
If you can't reason why, think about your brain and how it works for a minute. Just like artificial neural networks, it is also made from millions of neurons, which function only when there's communication between them. Since artificial neural networks try to imitate the human brain, they need to somehow replicate neurons in the brain and connections between them (weights). This association will be made less abstract throughout this section.
Today, artificial neural networks can be used to tackle any problem that regular machine learning algorithms can. In a nutshell, if you can solve a problem with linear or logistic regression, you can solve it with neural networks.
Before we can explore the complexity and inner workings of an...
