Deep Learning Dictionary - Lightweight Crash Course

Deep Learning - Deep Learning Dictionary

Deep learning is a subfield of machine learning that makes use of algorithms called artificial neural networks (ANNs) that are loosely inspired by the biological structure and function of the brain to learn from data.

We'll get a more formal introduction to ANNs later, but for now, we can simply think of these networks as structures that accept some input, do processing on that input, and then deliver an output.

This processing occurs over a series of layers in the network, and when the network has more than one internal or hidden layer, it's considered deep, which is where the deep in deep learning comes from.

We can use these networks for various tasks and objectives, like classification tasks, object detection, time series predictions, and much more.

The key to these networks is that we can train them to learn about a given dataset. This is where the learning in deep learning comes from.

This learning that takes can place can be broadly categorized as either supervised or unsupervised learning.

With supervised learning, the data that the network is trained on is labeled, whereas with unsupervised learning, the data is unlabeled. We'll touch on each of these categories in more depth later.