Understanding AI, Machine Learning and Deep Learning

[pallab_haldar]

In This blog I am going to discuss about AI, Machine Learning and Deep Learning. What does it mean by it. I have prepared a simple diagram that will tell you everything  what does it mean AI,ML and Deep learning.

AI : Replicate the Hunan intelligence  by creating labels (identifying features) and Targets to predict value from new data, perform task and take decisions.

ML:  In ML data scientist generally apply Algorithm to past data or a data pattern to predict, classify, and make decisions. Labels may be present (features) or may not be. Based on the data pattern, we can extract the label and target/predicted value objects.

Deep Learning : Using artificial neural networks and deep lanes in deep learning, we do multiple iterations and reduce errors by back propagating. Good for large amounts of data and can recognize complex patterns in pictures, text, sounds, etc. It synthesizes the labels (characteristics) from the dataset.

When Deep learning is beneficial over ML :

Deep learning has a  notable advantages over machine learning to handling unstructured data. Unlike machine learning algorithms that rely heavily on structured data inputs, deep learning models can effectively process unstructured data types such as images, speech, and text.

It create multiple Layers of Neurons an identify the error suing forward propagation and then backpropagate the error to each layers. It repeat the process for a number of times until the loss reduces and get the accurate result. 

In the next blog I am going to discuss about usage of each ML algorithm using different scenarios.