An illustrated and intuitive guide to Neural Networks

If you’ve got read my previous articles, you’ll know what’s coming next. On this a part of the web, we take complex-sounding concepts and make them fun and nbd by illustrating them. And if you happen to haven’t read my previous articles, I highly recommend you begin with my series of articles covering the fundamentals of machine learning since you’ll find that plenty of the fabric covered there’s relevant here.

Today, we’re going to tackle the large boy — an introduction to Neural Networks, a form of machine learning model. That is just the primary article in an entire series I plan on doing on Deep Learning. It would give attention to how an easy artificial neural network learns and give you a deep (ha, pun) understanding of how a neural network is constructed, neuron by neuron, which is super essential as we’ll proceed to construct upon this information. While we’ll dive into the mathematical details, there’s no have to worry because we’ll break down and illustrate each step. By the top of this text, you’ll realize that it’s waaaaay simpler than it sounds.

But before we explore that, you is perhaps wondering: Why do we’d like neural networks? With so many machine learning algorithms available, why select neural networks? The answers to this query are plentiful and extensively discussed, so we won’t delve too deeply into it. However it’s price noting that neural networks are incredibly powerful. They will discover complex patterns in data that classical algorithms may struggle with, tackle highly complex machine learning problems (reminiscent of natural language processing and image recognition), and diminish the necessity for extensive feature engineering and manual efforts.

But all that said, neural network problems just about boil all the way down to 2 principal categories — Classification, predicting a discrete label for a given input (ex: is that this an image of a cat or a dog? is that this movie review positive or negative?) or Regression, predicting a continuous value for a given input (ex: weather prediction).

Today we’ll give attention to a regression problem. Consider an easy scenario: we recently moved to a brand new city and are currently looking for a brand new home. Nonetheless, we notice that the costs of homes in the world vary significantly.

Since we’re unfamiliar with town, our only source of data is what we…