what is machine learning ? and what makes mechine more intelligent
Machine Learning:
It is a subset of a massive superset named Artificial Intelligence. Let’s know a little about it.
Artificial Intelligence:
What is Artificial Intelligence:
In simple words, it is defined as the science of creating intelligent machines which mimic human behaviour
What makes a machine intelligent?
It is just the same as what makes a person/student intelligent. In the present world scenario, the tests conducted in the institution is the criteria for intelligence. In the same way, there is a test called the Turing test to judge a machine’s intelligence.
Turing test:
In the Turing test, there will a person who asks random questions to both the human and the machine. The questionnaire has to differentiate the answers given by the human and the machine. If the questionnaire makes the correct guess for more than half of the answers, then the machine fails. In that case, the machine is not artificially intelligent.
Differentiating AI, ML, DL:
The above image differentiates AI, ML, DL. Machine Learning is a subset of Artificial Intelligence and Deep Learning is a subset of Machine Learning which mainly works on Neural Networks.
What is Machine Learning:
There is no perfect definition of machine learning. The two definitions which define machine learning are:
Arthur Samuel (1959). Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed.
Tom Mitchell (1998) Well-posed Learning Problem: A computer program is said to learn from experience E concerning some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.
This seems a little confusing. Let me detail it with an example.
The example of classifying emails as spam and not spam.
Experience:
Watching you label emails as spam or not spam.
Task:
Classifying emails as spam or not spam.
Performance:
The number (or fraction) of emails correctly classified as spam/not spam.
There are three subparts of machine learning...not spam are Supervised Learning, Unsupervised Learning and Reinforcement Learning.
Supervised learning:
In a supervised learning model, the algorithm learns on labelled spam are vary data set, providing an answer key that the algorithm can use to evaluate its accuracy on training data.
It is of two types:
Regression
Classification
1.Regression:
Regression analysis methods are used to predict a continuous value output.
Regression example:
The above example is the prediction of house price based on size. The house size and its price vary almost linearly in the above example.
Classification:
classification is a supervised learning concept that categorizes a set of data into classes.
In the above example, the algorithm predicts whether the image of a duck or rabbit. It names a set of animals like ducks and the other set as rabbits. This is an example of classification.
Unsupervised learning:
An unsupervised model, in contrast, provides unlabeled data that the algorithm tries to make sense of by extracting features and patterns on its own.
It is of two types:
1.clustering
2.dimensionality reduction
1.clustering:
Clustering is an unsupervised machine learning task that automatically divides the data into clusters or groups of similar items.
In the above example, the algorithm just differentiates the animals based on features and cluster them. The algorithm doesn’t name the clusters.
2.dimensionality reduction:
Dimensionality reduction refers to techniques that reduce the number of input variables in a dataset.
https://media.geeksforgeeks.org/wp-content/uploads/Dimensionality_Reduction_2.jpg
The above image is an example of dimensionality reduction. In the above image when the axis is rotated then the three-dimensional points are converted into two dimensional. In this way, we can reduce dimensions.
Reinforcement learning (RL):
It is an area of machine learning concerned with how software agents ought to take actions in an environment to maximize the notion of cumulative reward
In the above image, there is a baby who is moving in a room full of obstacles for its food. When the baby hits an obstacle he will cry or else he will reach his food. The action here is moving. The reward is food.
This is in brief about Machine Learning.
Super..
ReplyDelete