Machine Learning
Machine Learning (ML) is a branch of Artificial Intelligence that uses algorithms to learn from data, and, by noticing the patterns and trends in it, predict future data.
There are mainly 3 different types of ML. They are:
1. Reinforcement Learning:
Reinforcement Learning (RL) works on the principle of rewards. This basically means that when the model gives a prediction, if it is correct, the model is given a "reward". If it is incorrect, the model is not given a reward. In this way, the model learns from its mistakes.
Reinforcement learning models do not require human supervision.
Reinforcement learning is used in various fields like Robotics, Healthcare, Gaming, Natural Language Processing (NLP), Finance, etc.
Supervised Learning:
In supervised learning, the machine is supervised to check if it has got the correct output. We give labels to the training data set and the testing data set. In this way the computer learns what are the different types.
We can give names, for eg, if an object is rectangular, has four wheels, then it is a - Car. If it has 2 wheels it is a- Bike.
In this way, the machine learns from the training data set and uses it to predict the test data set.
Unsupervised Learning:
In unsupervised learning, the machine does not need human intervention. The machine is given raw data, which is neither labelled or classified. If we give images of 10000 cars and 10000 bikes, the computer learns where it should try to get the matching parts, for eg, it might recognise a car which has 4 wheels. If we give an image of a bike, for eg, it will find that it has pedals etc as its way to recognise the data if it is a bike or not.
Thank You
Aniruddha KP
References:
(1) Tech Vidvan (Fig. 2 + content)
(2) Geeks for Geeks
(3) Tech Vidvan (Fig. 1 + content)
Thumbnail credit: Express Analytics