Deep learning is a category of Machine Learning, a neural network with three or more layers (input layer, hidden layers, and output layer or multiple hidden layers). These neural networks mimic the behaviour of the human brain in matching the ability of machines and enabling them to learn. A single layer of a neural network makes rough predictions, whereas hidden layers can help get good accuracy. The idea followed by Neural Network is the biological neurones, which are nothing but brain cells. Deep learning is capable enough to focus on the accurate features themselves by requiring a little guidance from the programmer and is very helpful in solving the problem of dimensionality. Deep learning consists of a series of levels, and each level learns how to translate its input data into output data. It improves automation in performing analytical and physical tasks without the interference of humans. Deep learning technology can be utilised on day-to-day based tasks like image visualisations, voice-controlled assistants, self-driving, and automatic machine translation).
Deep Learning has Artificial Neural Networks (ANN) and Convolution Neural Networks (CNN)
Artificial Neural Networks (ANN): Artificial Neural network (ANN) is inspired by biological neural networks. In ANN, the network is interconnected by neurones which send messages to each other. The three main parts of ANN are the input layer, hidden layers and output layer.
Convolution Neural Networks (CNN): Convolution Neural Networks (CNN) are mainly used for the recognition, detection and classifications of images. CNN performs with the Convolution layer; it is the first layer to extract features from an input image, and then it connects to the Pooling Layer, which reduces the number of parameters when the photos are too large.
