In this 21st century with the development of sophisticated technology in all fields life has become easier compared to the previous decades. One such result obtained by the development of the sophisticated technology is Artificial intelligence, which we are all well familiar with, as it has been a popular focus in the movies like The Matrix, The terminator, Avengers etc. Lately we have been hearing about other terms like ''Machine Learning'' and ''Deep Learning'' which are used often when talking about Artificial Intelligence.
Though these terms are used interchangeably, they are different. There have been various stages of AI since its inception in 1950's. AI is a broad area of computer science which enables the computers to think in which Machine learning is one of the sub areas. ML is part of AI and is a bunch of statistical tools to learn from data.
Deep learning is also a sub area of Artificial Intelligence which has taken shape since 2006 and deals with neural networks and multi-layer neural networks. To implement all these Artificial Intelligence, Machine learning and deep learning many mathematical concepts like probability, statistics, Linear algebra, numerical optimization and computer programming are to be incorporated. Out of all the sub areas of Artificial intelligence, deep learning is the fast-evolving sub area whose implementation and research is being concentrated widely with the development of the programming languages like python and the powerful arsenal packages like Tensor flow, Theano etc. Let us see each of these topics and their applications in different fields in short and simple way.
The goal of AI is to create systems that can function independently and intelligently. Humans can speak, listen, and communicate through language. This is the field of speech recognition. Much of speech recognition is statistically based, hence is called statistical learning. Humans can write and read text in a language. This is field of Natural Language processing(NLP). Humans can see with their eyes and process what they see. This is computer vision. Humans recognize the scene around them through eyes which create images of the world. This field of image processing which even though is not directly related to AI is required for computer vision. Humans can understand their environment and move around fluidly, this is the field of robotics.
Humans have ability to recognize patterns and group things of like objects based on similarity. This is called pattern recognition. Machines are more capable than humans in pattern recognition by processing huge amounts of data and the dimensions of data. This is the field of Machine learning. Human brain is a network of neurons and use these neurons to understand and learn things. If we can replicate the structure and the function of the human brain we might be able to get cognitive capabilities in machines. This is the field of neural networks. If these networks are more complex and deeper and we use those to learn complex things that is the field of deep learning.
There are different types of deep learning in machines which are essentially different techniques to replicate what the human brain does. If we get the network to scan images from left to right, top to bottom then this neural network is a convolution neural network(CNN) which is used to recognize objects in a scene. This is how computer vision fits in and object recognition is accomplished through Artificial Intelligence. Humans have the capability to remember things. We can remember what we had for dinner last night, well at least most of us. We can get a Neural Networks to remember a limited past. This is a recurrent NN. There are 2 types Neural Networks. They are:
1. Symbolic based and
2. Data based
Data based Neural network is also called as machine learning in which we need to feed the machine lots of data before it can learn. For example, if we have lots of data for sales vs advertising spend we can plot that data to see some kind of pattern. If the machine can learn this pattern, then we can make predictions based on what the machine has learnt. While humans can easily learn in 1-dimension, 2-dimension and 3-dimensions, machines can learn in many more dimensions like even in 100 or 1000s of dimensions. Machines can look at lots of high dimension data and determine patterns. Once the machine learns these patterns, it can make predictions that humans can't even come close to. We can use all these machine learning techniques to perform either classification or prediction. For example, when we use some information about customers to assign new customers to a group like young adults then you are classifying that customer. If you are using the data to predict if they are likely to move on to the competitor, then you are making a prediction. There is another way to think about learning algorithms used for AI. If we train an algorithm with data that also contains the answer, then it is called supervised learning. For example, when we train a machine to recognize our friends by name we will need to identify them for the computer. If we train an algorithm with data where you want the machine to figure out the patterns, then it is unsupervised learning. For example, we might want to feed the data about celestial objects in the universe and expect the machine to come up with the patterns in the data by itself is an example of unsupervised learning. If we give any algorithm any goal and expect the machine to achieve that goal through trial and error then it is called reinforcement learning. A robot attempting to climb a wall is an example of reinforcement learning.
From the explanation and the applications stated above we can say that Machine learning and deep learning are branches of Artificial Intelligence which appear to be similar but are different from each other and each of these has diverse applications in the modern world. With the advancement in all these technologies, many difficult tasks have become simpler and will become even more simpler. Each of the sub areas of Artificial Intelligence is very vast and will take many years to do research, learn, implement and develop.