Introduction

To start learning artificial intelligence, one must have an idea about the difference between some of the most subsets of artificial intelligence, in this article, I will try to give you an idea about the difference between the most known subsets of artificial intelligence which are machine learning and deep learning. Fortunately, making the difference between these two subsets is not that difficult!

Requirements to follow this blog

There are no special requirements to read this blog! The only thing the reader must have is the curiosity to know more about artificial intelligence.

Notations & acronyms

• ML : Machine Learning

• DL : Deep Learning

• ANN : Artificial Neural Network

What is the difference ?

Machine learning is a set of algorithms to make computers learn from data to perform a task without being explicitly programmed. In classical programming, we need inputs and mathematical rules to produce the output. In contrast, in machine learning, we have the data and the algorithm must learn the rules.
Deep learning uses a complex structure of algorithms inspired by the human brain. This makes it possible to process unstructured data such as texts, images and signals.

What is machine learning ?

Machine learning is a set of algorithms that allow computers to learn without being explicitly programmed. In machine learning, the algorithms cause the computer to learn patterns in the data. The goal of machine learning algorithms is to be able to learn from a set of data called a training dataset and to be able to generalize on unseen data. As a machine learning developer, you need to be able to develop algorithms that can learn and recognize different patterns in a given dataset.
The learning process of machine learning can be supervised or unsupervised. In supervised learning, the data is annotated, which means that for each input (e.g. image), we have its output (e.g. label: person, cat, ...). In unsupervised learning, the data is not annotated and the aim of the algorithm is to find patterns based on the proximity of the data.

What is deep learning ?

Deep learning is a subset of machine learning that contains more complex algorithms inspired by the human brain. When dealing with unstructured data (such as images, text, signals, ...), simple machine learning algorithms are unable to extract the features needed to learn the patterns in the data. DL is based on an architecture called ANN (Artificial Neural Networks) that was originally inspired by human neurons and is capable of learning and extracting more difficult features in the training dataset.
In machine learning, we extract manually the necessary features from the data before feeding the algorithm, however, deep learning algorithms are capable of extracting these features.

To summarize ...

• In machine learning, we do not explicitly program computers.

• Deep learning is a subset of machine learning.

• Deep learning algorithms are based on architectures called ANNs (Artficial Neural Networks).

• Machine and deep learning need data to learn.