Source – indiatimes.com

The buzzword right now is artificial intelligence. Within artificial intelligence, the sub-sector that is gaining the most traction is deep learning. You hear it everywhere and the talk is that it will disrupt all major industries.

But, what really is Deep Learning?

Deep learning is a method of artificial intelligence which is used to help machines do something that is natural to humans- think and take decisions logically. Deep Learning methodologies can potentially help a self-driving car figure out traffic signs or maybe identify defects in a product manufacturing line.

Its work is inspired by how the brain functions and makes connections. We can link one thing to another. The idea of deep learning is to make machines to do just that – think. Deep learning aims to ‘understand’ data as a whole and not be confined to just one task.

What is the difference between Artificial Intelligence (AI), Machine Learning (ML) and Deep Learning?

To oversimplify it – Artificial Intelligence (AI) is a broad study area and its subset is machine learning and machine learning’s subset is deep learning. Machine learning is a method wherein machine learns how to read data.

“AI is the biggest, ML is smaller and deep learning is even smaller,” said Andrew Ng, AI evangelist and co-founder, Coursera. All of these terms are connected to one another but are not the same. Machine learning is the current technology state we are in.

The ‘deep’ part of Deep learning is when a system has become ‘intelligent’ enough to classify data more and more accurately and can learn better. More importantly, the system can learn from mistakes and enhance itself.

Where can all deep learning be potentially useful?

Deep Learning can be used in self-driving cars, the manufacturing sector, gaming like in Go (the game), precision medicine and various analysis.