What is Artificial Intelligence: Machine & Deep Learning

Understanding the latest developments in AI can be overwhelming. It seems as if every day there is a new news segment about self-driving cars, warehouse robots, and smart algorithms. Through the smog of confusion, the wider picture of AI can be boiled down to two core concepts: Machine Learning and Deep Learning. Frequently these two terms are thrown together like interchangeable buzzwords, but there are significant differences and they are important to understand.

 Examples of machine learning and deep learning are all around us – it’s how Netflix know the next show you’re going to watch or how Facebook know when it’s appropriate to tag you in a photo. 

Artificial Intelligence is intelligence demonstrated by machines, in contrast to the natural intelligence (NI) displayed by humans and other animals. In computer science, AI research is defined as the study of “intelligent agents”: any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals. You can see our breakdown of AI: AGI, ANI, and ASI here.

What is Machine Learning?

Machine learning is an approach to achieve Artificial Intelligence. Technically, an algorithm that parses data, learns from that data and then applies what it has learned to make better decisions without being explicitly programmed. This means that a single program, once created will be able to learn to do some intelligent activities. This is in contrast to purpose-built programs, where the code defines its behavior. 

From a mathematical standpoint, not all AI is machine learning BUT all machine learning is AI. And there are 4 methods of machine learning: 

• Supervised Learning
• Unsupervised Learning
• Semi-Supervised Learning
• Reinforcement Learning 

Deep Learning is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms. Deep learning is a technique for implementing machine learning that uses the premise of neural networks for the computer to learn through examples – like humans.

Deep learning involves feeding a computer system a lot of data which it can use to infer and make decisions related to other lots of data in the future. The data is fed through “neural networks” like machine learning – a construction of binary true/false questions or questions with numerical values which can help classify data that is pushed through it.

 Deep learning algorithms can take messy and unlabeled data such as video, images and, text and impose enough order as to make useful predictions. As such, Deep learning is being used in some really sophisticated applications, like self-driving cars which are learning to recognize obstacles and react to them appropriately. They have also been used to beat humans at Go and Chess.