How Artificial Neural Networks works in Deep Learning

mediumThis post was originally published by Ranjit Singh Rathore at Medium [AI]

Hey Everyone , back with another blog related to deep learning , In this blog we are going to learn core concept behind Deep Learning i.e Neural Networks.

In this blog we are going to covers some basics concept under Neural Networks :

What are Neural Networks

Neural Network are the computing systems which are mainly inspired by biological Neural networks that constitute animal brains.

An Artificial neural network is based on collection of connected units or nodes called Artificial neurons . Each connection can transmits the signal to other neurons . An artificial neuron that receives a signal then processes it and can signal neurons connected to it. The “signal” at a connection is a real number, and the output of each neuron is computed by some non-linear function of the sum of its inputs. The connections are called edges. Neurons and edges typically have a weights that adjusts as learning proceeds. The weight increases or decreases the strength of the signal at a connection. Neurons may have a threshold such that a signal is sent only if the aggregate signal crosses that threshold.

Neural Network Architecture.

Artificial Neural Network Architecture

In Artificial Neural Networks Architecture , we mainly have three layers :

Input Layers : The features are taken as input in the input layer ,and each neurons represent each feature from the data set. The number of neurons in this layer depends on the number of features provided in the data set .

Hidden Layers : The activity of each hidden unit is determined by the activities of the input units and the weights on the connections between the input and the hidden units. this also called hidden layer.Hidden layers can have multiple layers ,and each layers can have different number of neurons in it, depend on the performance of the network and according to that we can adjust them .

Output Layer : It is the last layer in the ANN architecture .The behavior of the output units depends on the activity of the hidden units and the weights between the hidden and output units. There can be any number of neurons in it ,based on the type of problem we are solving .

Different types of Neural Networks

There are mainly 3 types of Neural Networks :

Artificial Neural Network : It is an type of neural network ,which is a collection of connected artificial nodes called neurons. The main thing which differentiate it from other neural network types is the Input Format. In this the input feed into this network are in the numeric formats. Some application of ANN is in problem related to numeric data set and task of predicting the regression and classification tasks.

example of how ANN type networks works.
Applying ANN in MNIST Dataset in order to classify the digits.

Convolution Neural Networks:In deep Learning, a convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied to analyzing visual imagery . They are also known as shift invariant or space invariant artificial neural networks (SIANN), based on their shared-weights architecture and translation invariance characteristics . They have applications in image and video recognition, recommender systems, image classification ,medical image analysis , NLP ,etc.

Spread the word

This post was originally published by Ranjit Singh Rathore at Medium [AI]

Related posts