Machine Learning vs Deep Learning

At some point, most of us have struggled to understand the difference between ML and Deep Learning

What is Machine Learning?

It is a way of achieving artificial intelligence by using different kinds of techniques and algorithms including

• Support Vector Machines(SVMs)
• Decision trees
• Naive Bayes
• Linear regression
• Neural networks

and so on…

                       

What is Deep Learning?

It’s a subset of ML that specifically uses neural networks. Neural networks have evolved from ML techniques over time. They are much more effective and provide significantly better and more accurate results.

Why are other ML algorithms still used?

Neural networks(especially deep ones) require a lot of computational power, take longer to train, overfit in some cases and require huge amounts of data to function effectively.

Think of it this way, making use of neural networks is like using the quadratic formula to solve equations

                          

Its a lot cooler and makes you look like a pro, but it also requires more effort and is only recommended when we come across huge numbers.

Factoring, on the other hand, is a simple yet effective method which can be used for simple applications.

Moreover, there are cases where ML techniques have proved to be more effective than their deep learning counterparts.

Conclusion

Its better to use ML algorithms for simple tasks such as

• categorization of files/documents
• classification of news articles into Tech/Business/Politics etc
• spam filtering
• sentiment analysis(detecting positive, negative or neutral emotions from text)

On the other hand, deep learning algorithms can be used for relatively complicated tasks, when we have large amounts of training data at our disposal

• self driving vehicles
• face recognition softwares
• natural language processing
• virtual assistants
• ad marketing etc