EVERYTHING THAT YOU NEED TO KNOW ABOUT THE CONCEPT OF DEEP LEARNING

Have you ever wondered how Google Translates a whole webpage to a special language in a few seconds or your phone gallery groups images supported by their location? All of this is a product of deep learning.

But what exactly is deep learning?

A subset of machine learning, deep learning is also a subset of AI. AI permits machines to mimic human behaviour. Machine learning gives us the technique to realize AI through algorithms. Algorithms train with data and eventually develop a deep learning kind of machine learning inspired by the structure of the human brain in terms of deep learning. This structure is named a ‘man-made’ neural network.

Let us understand deep learning better and the way it is different from machine learning

If I create a machine differentiates between tomatoes and Cherry's using machine learning, we will need to tell the machine the features on which the 2 are often differentiated. These features might be Size and the sort of stem on them. But with deep learning, these differences are picked out by the neural network without human assistance or intervention. However, this sort of Independence is gained at the expense of training the machine with a huge volume of knowledge.

A dive into the working of neural networks

This is best illustrated with an example.

Let assume that there are three students. Each of them writes down a nine-digit on a paper. But the way we write numerals is very varied depending on our handwriting and each one of us had unique handwriting, but these differences are picked by a human bring easily and can read the same number written three different ways by people with 3 different handwriting. Now imagine if a computer had the same kind of knowledge, and this is where deep learning comes in. With deep learning, it is possible to create a neural network trained to spot handwritten digits.

Let us assume on a screen, each number represents a set of pixels. Now that totalled amounts to a much higher number of pixels neurons. At the core of a neural network the knowledge, processing takes place. Each of the pixels gets fed to a neuron within the First layer of our neural network. This forms the input layer on the opposite end. We have the output layer with each neuron representing a digit with a hidden layer existing between them. The knowledge is transferred from one layer to a different over connecting channels. Each of those features a value attached to them and hence is called a ‘weighted Channel’. All neurons have a singular number related to it called bias. This unique bias of each of the neurons gets added to the weighted sum of inputs. Inputs reaching the neuron are then applied to a function referred to as the ‘activation’ function, the results of the activation function determine if the neuron gets activated every time activated neuron passes on information to the subsequent layers.

This continues up until the second last layer, the one neuron activated within the output layer corresponds to the input digit. The weights and biases are continuously adjusted to supply well-trained work.

So How can Deep Learning help in Customer Support?

Increasingly now marketers want to create a seamless and consistent consumer experience. A part of this process is the automation of processes like answering simple consumer queries or addressing a visitor who has come to visit an online website or store. These are done by chatbots who can perform both voice and text interactions. The banking industry calls it ‘Conversational banking’.

Many consumers do not realize that most customer support agents conversing with them, while it is seeming real are not. They do not even realize that it's a bot on the opposite side in medical aid neural networks that detects the cancer cells and analyses MRI images to offer detailed results. Self-driving cars which appeared to be a fantasy a decade ago is now a reality. Apple, Tesla and Nissan are only some of the companies creating self-driving cars.

Deep learning features a vast scope but on the face it, it has some limitations.

The primary one as we discussed earlier is data needed for deep learning is that the most effective. Unstructured data in a neural network requires a huge volume of knowledge to coach. Let's assume that we have access to the required amount of knowledge processing, but it is often not within the potential of each machine which brings us to our second limitation.

Gaming computational power training a neural network requires graphical processing units which have thousands of cores as compared to CPUs and GPU, these are very expensive. Finally, the time deep neural networks take hours or maybe months to coach. This learning time increases with the quantity of knowledge and the number of layers within the network.

AI and Microsoft cognitive toolkit are considering the long-term future for deep learning and AI. We have only scratched the surface. Technology can exist that perform functions for the blind using deep learning with computer vision to explain the planet to the users replicating the human mind at everything. Maybe AI enabling our lives with will not be a fantasy for too long. The future is filled with surprises many of which will come from deep learning.