Deep Learning vs Machine Learning
Posted By Rajat Kukrety | 28-Feb-2018
What is Machine Learning?
According to Tom Mitchell "A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E".
This can be more clear with real-life examples as the prediction of body weight and height.
Let's assume we have to build a software that predicts the person's weight based on his height. We can collect some samples which have both height and weight of few number of persons and then based on their data machine can predict the actual height of person for some random weight. Also, the machine can identify it based on the region of person or gender to be more specific.
Whereas Deep Learning has been around for a couple of years. Deep learning is more popular than Machine Learning nowadays.
What is Deep Learning?
Deep learning is a particular kind of machine learning that has the flexibility to learn and represent the world as the nested hierarchy of concepts and more abstract representations computed in terms of less abstract ones.
To make it more clear let's take an example of a system which has to recognize the image has Dog or Cat. On a normal approach, anyone would distinguish that it has whiskers or not or it has pointed ears or not. We will feed the facial features in it and let the system identify in classifying a particular animal that which features are more important but Deep Learning is one step ahead. It automatically judges which features are more important to classify it and in ML we have to give it manually.
These are some areas where we can apply them
Information Retrieval: for applications like image search, both text search, and search engines.
Computer Vision: for applications like and facial recognition and vehicle number plate identification.
Marketing: for applications like target identification, automated email marketing.
Medical Diagnosis: for applications like anomaly detection, cancer identification.
Natural Language Processing: for applications like photo tagging.