Deep Learning: Types and Its Applications

1 minute read

Published:

Deep Learning is one of the advanced fields of this decade that revolutionized Artificial Intelligence as a whole. It is deep learning that created the possibility of solving large-scale problems in the present world. Deep Learning is a branch of traditional Machine Learning where we extend the structure and functionality of Neural Networks to solve complex problems and model vast data for accurate predictions.

Due to the emergence of this field, researchers all over the world started exploring it. Amazon created Alexa, Google made search and Language Translation better, Facebook improved their image tagging more accurate — all are using Deep Learning behind the scene. Deep Learning is even integrated to other parts of A.I. as well such as Reinforcement Learning.

There are various types of Deep Learning models and their usage varies on factors like type of dataset, type of problem to solve. One might get frustrated while thinking about the starting point of Deep Learning. If you know basics of Machine Learning and Mathematics then you can simply get into the learning phase. I covered few projects as part of my Online Nanodegree and they are as follows:

ProblemDeep Learning ModelGithub Repo
Image Classification Using Deep LearningCNNHere
Generate TV Scripts Using Deep LearningLSTMHere
Language Translation Using-Deep LearningLSTMHere
Generate Faces Using Deep LearningGANsHere

You can easily follow instruction of Python and Environmental Setup to run Jupyter Notebook in each project. By going through these applications one can grap the intuition of how each model works.