Nowadays many misunderstandings are there related to the words machine learning, deep learning and artificial intelligence(AI), most of the people think all these things are same whenever they hear the word AI, they directly link that word to machine learning or vice versa, well yes, these things are correlated to each other but not the exactly same. Let’s bust these myths.
Just doing a simple Google search of these terms will return hundreds of Venn-diagrams trying to describe how these three terms are interrelated. Here's one such example, and it is difficult and confusing to understand some of these diagrams.
There's just way too much going on here to understand the certain relationships that we're trying to understand. What I'm hoping to lay-out in the In this blog is a much more simplified explanation of how these terms are related and how they are different from each other.
Machine learning is fitting a function to examples and using that function to generalize and make predictions about new and unseen examples. And then we can also conclude a simpler related definition that is just simple pattern matching. So a model will basically learn a pattern from past examples and then it will fit a function to it and then it'll use that function to pick up on similar patterns in future data to make a prediction about that data. So let this circle below represent all things machine learning. More detailed explanation is available here
Machine Learning Example:
If you are getting better product choice recommendations On Amazon, or better Tv series or movie or shows recommendation on Netflix this is all because of the fact that these big companies are using machine Learning For this purpose.
Machine learning contains Deep learning entirely within itself. That is, Deep Learning is a subset of machine learning.
Deep Learning is still fitting a function to examples, but the difference comes in the fact that these functions are then organized together as connected layers of nodes, sometimes also known as Neural networks. In other words, you'll have many functions connected together in one network where each function is responsible for a very specific task. Similar to Machine Learning, the goal of Deep learning is to generalize and make predictions about new examples using our network of functions. An alternative definition is just more advanced connected pattern matching. So again, instead of just using one function to match a pattern, you're just using many functions to match a pattern, where each function has a specific task in pattern matching. The one important thing to take away is just that deep learning is a type of machine learning. It is a subset of machine learning. There are things that are machine learning but not deep learning, but there's nothing that is deep learning but not machine learning.
Machine learning models become better progressively but the model still needs some Human guidance.If a machine learning model returns an inaccurate prediction then the programmer needs to fix that problem explicitly by doing Feature Engineering or tweaking other Parameters, but in the case of deep learning, the model does it by himself, this is the major factor to be considered in machine learning vs deep learning. Automatic car driving system is a good example of deep learning they learn to take decisions solely based on training data.
Deep Learning Example:
Suppose we have a Radio and we teach a machine learning model that whenever someone says “play” the radio should be on, now the machine learning model will analyze different phrases said by people and it will search for the word “play” and as the word comes the Radio will be on but what if someone said “I am very tired and want to listen to some music”, in this case, the user wants the radio to be on but the sentence does not the consist the word “play” so the Radio will not be on. That’s where deep learning differs from machine learning. If it were a deep learning model it would on the Radio, a deep learning model can learn from its own method of computing. Every voice assistant like Siri, Google home, Alexa are examples of deep learning
Now the last and most widespread term that we're going to discuss is artificial intelligence. Artificial intelligence or AI is a superset of machine learning and deep learning and contains almost everything within itself.
In defining it, we're going to divide AI into two separate definitions:
Types of AI:
Weak AI is intelligence specifically designed to focus on a very narrow task like spam detection in the mail or calculating the price of a House. The key take away in machine learning vs ai is that, weak AI is similar to machine learning, intelligence designed to focus on a very narrow task, but then you're not going to take that model that was built to detect fraudulent credit card charges and recommend Amazon products to you.
Strong or general AI is a machine with consciousness, sentience, and a mind. General intelligence is capable of any and all cognitive functions and reasoning that a human can do. So general AI is a superset of so many things. Only one of those things happens to be machine learning, but there are many things under the umbrella of AI that is not machine learning.
Example of AI:
The truth is we are not able to establish a proper general AI till now but we are far away to establish it. The reason we are not able to establish general AI till now is, we don’t know the many aspects of the human brain till now like why do we dream, How we learn, how we connect one skill learned in one task to another task? etc. Sad but true.
So this is just a high-level overview of how these three terms are related to each other. Deep learning is a subset of machine learning which is a subset of general artificial intelligence. They should not be used interchangeably without specifying what type of AI we're talking about.