What is Machine Learning?

What is Machine Learning?
Machine Learning can be defined as the study of computer algorithms that improve automatically through experience.
It is the process of making machine learn human activities and mimic them in their environment. Machine Learning is
considered to be a part of Artificial Intelligence. Machine learning algorithms build a model based on sample data, known as “training data”, in order to make predictions or decisions without being explicitly programmed to do so.

How it works?
Consider you are given a puzzle where you have to move a character from one room to another in a building where certain rooms are considered good and certain are considered bad. Now you being a human will understand the problem and you make faster decisions to move the character. But that’s not the case with machines, they are not that smart to make decisions on the first go. We have to train them based on various data we create. This data can be the various human gameplays and their moves which are recorded and passed on to the model. Now what happens is that the computer analyse this data and learns, this data is called training data . Now the trained model is tested to know how accurate it is? It is tested with data known as as test data. This helps us in knowing how precise the model is and help in improvement where required.

Machine learning approaches

Machine learning approaches are traditionally divided into three broad categories:

1. Supervised Learning
2. Unsupervised Learning
3. Reinforcement Learning

Supervised Learning:

It is a type of machine learning where there is a function that maps an input to an output based on example input-output pairs. In this we use labelled data to train the model. There are two important techniques of supervised learning:

1. Classification
2. Regression

Unsupervised Learning:

Unsupervised learning is a type of machine learning that utilizes a data set with no pre-existing labels with a minimum of human supervision, often for the purpose of searching for previously undetected patterns. Unsupervised learning classified into two categories of algorithms:

1. Clustering
2. Association

Reinforcement Learning:

Reinforcement learning is an area of machine learning concerned with how intelligent agents ought to take actions in an environment in order to maximize the notion of cumulative reward. In this the agent is trained in such a way that its actions are targeted on maximizing its reward.

Main parts in reinforcement learning:

1. Agent
2. Environment
3. Reward
4. Action
5. Policy
6. Value
7. Action-Value

Deep Learning:

Deep learning is a branch of machine learning which is completely based on artificial neural networks, as neural network is going to mimic the human brain so deep learning is also a kind of mimic of human brain.

Architectures:

▪Deep Neural Network — It is a neural network with a certain level of complexity (having multiple hidden layers in between input and output layers) . They are capable of modelling and processing non -linear relationships .

▪Deep Belief Network(DBN) — It is a class of Deep Neural Network . It is multi -layer belief networks .

Applications of ML:

Machine learning is a buzzword for today’s technology, and it is growing very rapidly day by day. We are using machine learning in our daily life even without knowing it such as Google Maps, Google assistant, Alexa, etc. Below are some most trending real-world applications of Machine Learning:

1. Image Recognition

2.Speech Recognition

3.Self-driving cars

4.Email Spam and Malware Filtering

5.Virtual Personal Assistant

6.Traffic prediction

7.Product recommendations

These are just a few applications of ML in today world. There are many more applications which are changing the world.

Limitations:

Although machine learning has been transformative in some fields, machine-learning programs often fail to deliver expected results. Reasons for this are numerous: lack of (suitable) data, lack of access to the data, data bias, privacy problems, badly chosen tasks and algorithms, wrong tools and people, lack of resources, and evaluation problems. There are also certain incidents where ML have gone wrong like In 2018, a self-driving car from Uber failed to detect a pedestrian, who was killed after a collision. Thus these machines need to be improved for better user service.

An Article by R.Vankata Varadhan