rasa.io better newsletter sends: https://rasa.io/?utm_medium=paid_display&utm_source=maii&utm_campaign=start-sending
What Is Deep Learning?
Blog Feature

By: Mike Kaput

Print this Page

April 28th, 2021

What Is Deep Learning?

We've answered the questions What is AI? and What is machine learning?

How about "deep learning"?

You'll hear this term referenced as you seek to understand, pilot, and scale AI, so it's important to understand.

Deep learning is the ability to teach the machine to think like the human brain. It's an advanced subset of machine learning. 

So deep learning is trying to mimic how the brain processes things, how it sees and understands things, and how it forms words and speaks.

You hear deep learning mentioned a lot in autonomous vehicles, natural language understanding, or computer vision. These are all areas where the machine displays human-like abilities to see, hear, and speak.

Deep learning systems use neural networks to achieve these capabilities. These are a series of algorithms that process information through various layers. Each layer is often in charge of a narrow task. Each successive layer builds on the previous narrow tasks. With enough layers optimized in the right way, you can give a machine impressive capabilities.

Take, for instance, a deep learning system that recognizes objects in photos. It uses complex layers in the form of a neural network to do what humans can do.

One layer might identify the edges of pixels around the object. Another might look at the context of the object's environment—is it outside or inside? Another might identify features that indicate if the object is animate or inanimate. And so on. The deep learning system then uses this data to predict what the object is.

The same thing happens with voice assistants like Siri and Alexa. The machine hears your voice, uses deep learning powered by layered neural networks to process all the various aspects of what you're saying, and tries to understand what you're saying or asking. It's trying to predict what you're saying or asking, so it can give you the proper response.

This type of system must be trained, like any machine learning technology. 

When identifying objects, a deep learning system must see millions of pictures. Humans must tell it which pictures represent which objects. With enough of this training, the machine can learn to identify objects independently, without humans.

Deep learning mimics the human brain, but it doesn't necessarily operate like it. You don't need to train a human toddler on millions of examples to train it how to identify a dog. With a relatively limited number of examples, the child can identify a dog with a high degree of accuracy.

Because of this, deep learning is human-like, but it doesn't operate like a human. It does what we can do, but differently than we do.

However, for practical purposes, it doesn't matter. Deep learning is advancing at a significant rate and powers many of the advanced AI applications available today. 

In the process, it's transforming business and marketing—so it's worth understanding for any company looking to maintain a competitive edge.

How to Get Started with AI and Deep Learning

Are you a marketer, salesperson, or business leader who wants to better understand how deep learning and other AI technologies can increase revenue and reduce costs?

You need our Ultimate Beginner's Guide to AI in Marketing

It has more than 100 resources to get you started with AI—without spending months on research or struggling with difficult jargon.

It contains our best resources for learning AI, the stuff that made the cut after we did 1,000s of hours of research and 100s of interviews. No hype. No fluff. Just the most actionable, relevant information to your company and career.

To get it instantly for free, click here—or click the image below.

New call-to-action

About Mike Kaput

Mike Kaput is Chief Content Officer at Marketing AI Institute and a senior consultant at PR 20/20. He writes and speaks about how marketers can understand, adopt, and pilot artificial intelligence to increase revenue and reduce costs. Full bio.

  • Connect with Mike Kaput
Disclosure: Marketing AI Institute writes about and recommends AI-powered marketing and sales technology. In all cases, content and recommendations are independent and objective. In some cases, Marketing AI Institute may have business relationships with companies mentioned, which may include financial compensation, affiliate compensation, or payment in kind for products or services. View a list of Institute partners here and MAICON sponsors here.