MACHINE LEARNING

Machine learning(ML) is a subset of Artificial intelligence. Machine learning algorithms build a mathematical data based on sample data or “training data” in order to make predictions without being programmed to do so. (Now you all might be thinking “Is this guy speaking English?!!” 😅😅). This basic definition may not be understandable because to understand this you need to know about the machine learning types.

ML TYPES :

• SUPERVISED LEARNING : In supervised learning the machine is provided with training data which is well labeled. As the name indicates this type of learning requires a supervisor who provides the machine with training data and trains it.

EXAMPLE:

Let’s take an example where the machine needs to classify whether a person is healthy or unhealthy. You, as a programmer would train the machine using various data like whether the person has fever,cough or cold. If he has any one of these then you would train the machine to classify the person as “unhealthy person”and vice versa.

• UNSUPERVISED LEARNING: In this case, the data is you give is not labeled and you are unsure about the output of the model. You model the algorithms in such a way that the machine can understand different patterns from the data and you do not interfere when the algorithm learns.

EXAMPLE: Let’s take an example where a machine has to categorize idly and dosa from a cluster of idlies and dosas in a plate. You do not need to explicitly train the machine on how idlies and dosas looks like for this. You can just make the machine understand different patterns. Now the machine won’t know what is dosa and idly but it knows different patterns like shapes, design ,colour or whatever. So based on these patterns the machine will be able to separate the two dishes.

• REINFORCEMENT LEARNING : It’s a kind of iterative process where the algorithm is executed sequentially based on trial and error. If the sequence maximises the performance then the learning is said to be positive reinforcement and vice versa.

EXAMPLE: Let’s take an example of a baby which doesn’t know how to walk. Now what it does is it observes it’s parents and others and see how they all walk. Then, it tries to replicate them and in the process it undergoes lot of hurdles ,then improves and then it finally learns to walk. This is exactly what happens in reinforcement learning. The machine tries to learn by observing its environment.

This is just a basic overview of machine learning. Actually,its a huge ocean out there but for an understanding I hope this would be enough.

You think machine learning is easy ?!! Do you know it takes nearly 4 months for a system to learn the location of living room,bedroom,kitchen and other parts of your house? Now you can see what kind of an idiot a machine is but it surpasses humans in speed and computational power.

So next time when your teacher calls you a stupid,tell him/her,”Atleast i am better than a machine”but make sure you dont drag my name while saying that.😂😂