In Today's world it is impossible to explain artificial intelligence, even if you hear the experts, they will define AI very loosely, reason being, Intelligence has no standard globally accepted definition
I plan to cover here the basic terminologies to get the basic building blocks right, Here we go
1. The field of Artificial Intelligence can be divided into two categories - ANI and AGI. Today we can see a lot of use cases of ANI being developed and used in the real world, on the other hand many researchers are working on ANI, we are still a long way away from it, but the closest to it is autonomous cars.
2. All Deep Learning is Machine Learning but All Machine Learning is not Deep Learning
3. Machine Learning : is a science of getting the computer to act without being explicitly programmed (Stanford University). To simplify this, ML is a technique which can analyse patterns and trends in large complex data which helps people solve the following type of problems
Classification : Identifying the email is SPAM or Not
Clustering: Personalisation based on user clusters
Regression: Credit Rating based Loan approval
Dimension Reduction : Visualising big data
4. Learning Techniques
To equip a machine the capability to analyse the data, the machine needs to be trained, similar to the way, a child is taught to do multiple things. I will continue with example of a child to explain the 3 learning techniques.
Supervised Learning: (Training the machine on a set of data that contains both the inputs and the desired outputs) When you are introducing fruits to child, you will show them an orange and tell them its an orange, but if you do this only one time, it may happen that the child may get confused between a large lemon and an Orange, but if show the child, 10 types of orange, the chance of the child getting Confused is Less.
Unsupervised Learning (Training the the machine with a set of data that contains only inputs and no desired output labels)
When you are introducing a child to fruits and you multiple types of fruits in front of a child and just ask them to categorise. They may categorise the fruits based on their color or shape or size. They may put pears and apple together considering the shape.
Reinforcement Learning (Active Learning based on feedback form of positive or negative reinforcement in a dynamic environment)
Again, when you are introducing a child to fruits, if you give an apple to the child and the child eats the whole apple, along with seeds , the child will understand that seeds should not be eaten (negative feedback) when having an apple
I think I have covered enough for one blog to be read and understood, Since there is still a lot to cover and if you are not feeling dazed , then head on to part 2 - https://www.aimadesimple.com/post/ai-terminologies-explained-simply-2