NEURAL NETWORK

NEURAL NETWORK

WHAT IS NEURAL NETWORK?

     

                             In information technology, a neural network is a system. of hardware and/or software patterned after the operation of neurons in the human brain. Neural networks also called artificial neural network -are a variety of deep learning technology, which also falls under the umbrella of artificial intelligence, A.I.

     .      Commercial application of these technologies generally focuses on solving complex signal processing or pattern recognition problems.

                              A neural network is a network or circuit of neurons, or in a modern sense, an artificial neural network, composed of artificial neurons or nodes. Thus a neural network made up of artificial neural network solve artificial intelligence problems. These artificial networks may be used for predictive modelling, adaptive control and application can be trained by database .self-learning resulting from experience can occur within networks, which can derive a conclusion from a complex and seemingly unrelated set of information.

HISTORY OF NEURAL NETWORK:-

                           The history of the neural network goes back to the early days of computing in 1943, mathematicians Warren MC Culoch and Walter Pitts built a circuitry system intended to approximate the functioning of the human brain that ran simple algorithms.

                             In the late 1940's psychologist, Donald Hulb created a hypothesis of learning based on the mechanisms of neural plasticity that is known as HEBBIAN LEARNING. Hebbian learning is considered to be 'typical' unsupervised learning rule and its later variants were early models for long term potentiation.

                            Franky and claek(1954) first used computational machines, then called calculators, simulate a Hebbian network at MIT. Other neural network computational machines were created by Rochester, Holland, Habit and Duda(1956).

                           In 1969, MIT researcher Marvin Minsky and Seymour Papert published the book "perceptions", which spelt out several issues related to the neural network.

                          It was not until around 2010 that research picked up again. The big data trend, where computers were able to compute large data sets .in2012 a neural network was able to beat human performance at an image net competition.

                                        Nowadays neural network technologies are growing rapidly. Artificial personal assistance like google home, Alexa is the product of neural networks. Now, these networks can predict weather condition, predict your needs, recommended products for us and also can carry out specific works for us like washing clothes to driving a car without the help of a human.

HOW NEURAL NETWORKS WORK?

                                   A neural network usually involves a large number of processors operating in parallel and arranged in tires. The first tire receives the raw input information- analogous to optic nerves in human visual processing. Each successive tier receives the output from the tire preceding to it, rather than from the raw input - in the same way, neurons further from the optic nerve receive a signal from those closest to it. The last tire produce output of the system.

                                  An artificial neural network involves a network of simple processing elements and element parameters. Aneural network contains layers of interconnected nodes. Each node is perception and similar to multiple linear regression. The perception feeds the signal produced by multiple linear regression into an activation function that may be non-linear.

 .                                  In a multi-layered perceptron (M.L.P.), are arranged in interconnected layers.

1- INPUT LAYER

2-HIDDEN LAYER 

3- OUTPUT LAYER

   

 INPUT LAYER:- The input layer collects input patterns. It collects a lot of information from the environment to produce the desired output.

OUTPUT LAYER:- The output layer has classifications or output signals to which input patterns may map. For instance, the patterns may comprise a list of quantities for technical indicators about security; potential outputs could be "buy", "hold", "sell".

HIDDEN LAYER:- Hidden layers fine-tune the input weightings until the networks margin of error is minimal. It is hypothesized that hidden layers extrapolate salient features in the input data that have predictive power regarding the outputs. This describes feature extraction, which accomplishes a utility similar to statistical techniques such as principal component analysis.

                                   Neural networks are notable for being adaptive, which means they modify themselves as they learn from initial training and subsequent runs provide more information about the world. The most basic learning models centred on weighting the input streams, which is how each node weights the importance of input from each of the predecessors.

HOW NEURAL NETWORKS LEARN?.

                            Typically, a neural network is initially trained or fed a large amount of data. Training consists of providing input and telling the network what suppose to be the output. These include gradient-based training, Fuzzy logic, genetic algorithms and Bayesian methods.

There are three types of learning---

SUPERVISED LEARNING:-THE LEARNING ALGORITHM IS GIVEN LABELED DATA AND THE DESIRED OUTPUT. FOR EXAMPLE, PICTURE OF DOGS LABELLED "DOG" WILL HELP THE ALGORITHM IDENTIFY THE RULES TO CLASSIFY PICTURES OF DOGS

UNSUPERVISED LEARNING:- THE DATA GIVEN TO THE LEARNING ALGORITHM IS ASKED TO IDENTIFY PATTERNS IN THE INPUT DATA. FOR EXAMPLE, THE RECOMMENDATION SYSTEM OF AN E-COMMERCE WEBSITE DISCOVERS SIMILAR ITEMS THAT YOU BOUGHT EARLIER. LIKE AMAZON, FLIPKART.

REINFORCED LEARNING:-THE ALGORITHM INTERACTS WITH DYNAMIC ENVIRONMENT THAT PROVIDES FEEDBACK IN TR=ERMS OF REWARDS AND PUNISHMENTS, FOR EXAMPLE, SELF DRIVING CARS BEING REWARDED TO STAY ON THE ROAD.

TYPES OF NEURAL NETWORK:- 

                             Neural networks are classified in how many layers they have between input and output.

1. The simplest varrient is the feed-forward neural network. This type of artificial neural network algorithm passes information straight through from input to processing nodes to outputs. It may or may not have hidden node layers.

2. More complex is recurrent neural networks. These deep learning algorithms save the output of processing nodes and feed the result back into the model. This is how the model is said to learn.

3. Convolutional neural networks are popular today. Particularly in the case of image recognition. This specific type of neural network algorithm has been used in many of the most advanced applications of AI including facial recognition, text digitization and natural language processing.

APPLICATION OF NEURAL NETWORK:-

          .           Image recognition was one of the first areas to which neural networks were successfully applied, but the technology has expanded to many more areas.

                                             A neural network evaluates price data and unearths opportunities for making trade decisions based on the data analysis. The network can distinguish subtle non-liniar interdependencies and patterns other methods of technical analysis can not.

                                          Some key uses of the neural network are as follows

1.Chatbots

2.Natural language processing, translation and language generation.

3.Stick market prediction

4.Weather prediction

5.Delivery driver route optimization.

4.Drug discovery and development.

5.Detecting disease and cure it 

6. Provide personal assistance.

7. Reach to the parts where a human can not reach.

CONCLUSION:-

                                     Overall, we can say that Artificial neural networks are developing day by day and in the near future, these networks will be a helping hand of man and mankind.