Machine learning is a type of artificial intelligence (AI) that allows computer to learn from data (image, data, text, speech, video). Researcher used deep learning has a great succeed at classifying data. That's why, this method is so popular in the last decade. Deep learning is also known as Convolutional Neural Networks (CNN). CNN was inspired by human brain nervous system that could classify objects they see or decide an action by learning data. CNN was introduced by
Kunihiko Fukushima in 1980 and modified by
Yann Lecun in 1998. This CNN could automate human task and help them efficiently.
 |
| Fig. 1. Neural Networks Nervous System |
CNN contains three basic components, such as convolution layer, activation function, and max pooling layer. Below is the picture of CNN architecture named AlexNet.
 |
| Fig. 2. AlexNet CNN Architecture |
So, what are the functions of this complex method and what are the real implementation of CNN?
- Nvidia is building and testing the autonomous car (without driver). This car has high technology sensors to avoid accidents.
- Google has developed google translate with speech recognition.
