Deep learning (DL), along with artificial intelligence (AI) and machine learning (ML), is poised to bring about a whole lot of changes to almost every industry. In this article in ITChronicles.com, Terry Brown explains the differences between each of these technologies, while also providing more insight into deep learning.
AI vs ML vs DL
Often used interchangeably, each of these terms means different things. Let’s understand these differences in detail below.
AI is a broader and more general term than ML and DL. As such, it involves the building of machines that can perform tasks characteristic of human intelligence. Typically, this includes things like planning, problem-solving, understanding languages, and recognizing voices and images. There are two types of AI, namely Narrow AI and General AI.
ML is a clearly defined term and refers to a specific subset of AI algorithms. As the name suggests, ML concerns itself with building machines powered with the ability to learn from data, and progressively improve on the task. ML enables ‘learning’ by feeding algorithms with data, so it can improve without human intervention.
DL is a branch of Machine Learning and takes ML to a whole new level. As such, it is a set of algorithms that endeavor to build a hierarchical representation of data. In this case, ‘hierarchical’ means that the algorithms work in ‘layers’ of functions, built upon one another. By operating in such a manner, they learn how to make accurate decisions. Furthermore, DL neural networks can decide which features are important to a task without human intervention.
DL in Practice
The use of DL is becoming more common over the past few years. While Facebook and Instagram, use it for image classification, LinkedIn uses its detect spam, abusive comments and to better understand the content.
That aside, DilogTech uses neural networks to classify inbound calls, while Netflix uses it for content recommendations. DL can also be used for self-driving cars and language translation and presents several possibilities for the healthcare segment as well. In consumer industries, DL can help reduce waste and optimize targeting, while in supply chain and manufacturing industries, it can help improve predictive maintenance of equipment, yield optimization, procurement analytics, and inventory optimization.
Click on the following to view the original article in full. https://www.itchronicles.com/technology/a-quick-guide-to-deep-learning/