Skip to main content
Blog eCommerce

An Introduction to Deep Learning

Bharat Kulkarni
August 23, 2019 |

Deep Learning - Computers

“O Deep Thought computer,” he said, “the task we have designed for you to perform is this. We want you to tell us…” he paused, “The Answer.”

—Douglas Adams, The Hitchhiker’s Guide to the Galaxy

The Inception of AI

In 1950 English Mathematician Alan Turing published a paper titled “Computing Machinery and Intelligence”, which opened the doors to the field that would then be called AI. And that spark of an idea is where it all started, but it is now the time where we have the quality of hardware and quantity of data that enables us to get closer to the uncertain future of AI.   

What is this buzzword all about?

Deep Learning is the subset of Machine Learning, which is a field dedicated to the study of algorithms and statistical models that provide systems the ability to automatically learn and improve from experience without being explicitly programmed.

Deep learning attempts to model high-level abstractions in data inspired by the structure and function of the brain called artificial neural networks.

In layman’s term, Deep Learning imitates the working of our brain. Like our brain has millions of neurons in a hierarchy and network of neurons that are interconnected with each other, that passes Electrical signals from one layer to another. This is how we humans learn things. Whenever we see, hear, feel and think something an electrical impulse is fired from one neuron to another in the hierarchy which enables us to learn, remember and memorize things in our daily lives.

In industry, deep learning is used to solve practical tasks in a variety of fields such as computer vision (image), natural language processing (text), and automatic speech recognition (audio).

Artificial Intelligence vs Machine Learning vs Deep LearningAI vs ML vs DL

As you can see in the above image, DL is a subset of ML, which is also a subset of AI.AI is the wide-ranging concept that was initially erupted, then followed by ML that thrived later, and lastly DL that is promising to escalate the advances of AI to the next level.

Artificial Intelligence is a very broad term that aims to bring out intelligent behavior in machines, “Intelligent Behavior” can be viewed from different perspectives, Machine Learning considers “a field of study that gives computers the ability to learn without being explicitly programmed.”

In simpler terms, feed data to a machine and let it learn by itself, as opposed to learning from rules or logical induction, which was heavily used at the beginning of Artificial Intelligence. Machine learning is a subfield of Artificial Intelligence and Deep Learning is a subfield of Machine learning

DL algorithms are inspired by information processing patterns found in a human brain. Instead of explicitly telling an algorithm which part or feature of data is important, we let the machine decide for itself. This is one of the reasons which helps Deep Learning algorithms achieve the state of the art results. AI vs ML vs DLIf you still don’t get it, don’t worry. Most of us don’t either.

These images below will make sense.

Data Science - Deep Learning

No one cared when it was called Regression. AI vs ML vs DL

It’s the same thing, but different words.

What value does Deep Learning add to eCommerce?

  1. Virtual Assistance –  Whether it’s Alexa, Google, or Siri, all of them use Deep Learning to provide hands-free service to order or know the status of their transactions.
  2. Chatbots and service bots – To provide a hassle-free customer service, a lot of companies use deep learning through which they can respond in an intelligent and helpful way.
  3. Personalized shopping/Product Recommendation – Amazon’s “Customers who bought this” and “You might also be interested in” that comes up with a list of items that you probably should be buying, those suggestions are exactly what you need, but you just never knew it before, right?
  4. Shopping Lens  – Saw a product but don’t know what it is? Don’t bother describing the product to google, just take a picture and you’ll be able to order it without any effort.

    Expect The Unexpected

    1. This resume does not exist – This DNN creates a resume that doesn’t exist. Deep Neural Networks can create something new that doesn’t exist but looks so real that we humans accept it without any second thought.
    2. Replace Objects – It’s possible to replace objects or interpret it as someone else in the video or image using Neural Networks, which is also known as DeepFake. Deep Learning 101
    3. The left image shows graffiti on a Stop sign, something that most humans would not think is suspicious. The right image shows a duck tape applied to a Stop sign. But unfortunately, the systems classify the sign on the right as a Speed Limit: 45 mph. Deep Learning 101
    4. Facebook shut down an experiment after they found out two AI chatbots to be chatting with each other in their own language. They created their own changes to the English Language that made it easier for them to work with, but humans couldn’t understand it.

Where are we with AI?

“We can only see a short distance ahead, but we can see plenty there that needs to be done.” – Alan TuringWe hope that we didn’t include a ton of technical jargon, if so then enjoy this one last comic.Deep Learning 101

Bharat Kulkarni

Bharat Kulkarni, a Data Scientist at DCKAP, loves working with Python. His interest lies in solving problems. Apart from being an avid reader, he is a cinephile, audiophile, and a quodophile. His hobbies include playing the guitar and ukulele, along with cycling and skateboarding.

More posts by Bharat Kulkarni