© 2016, IRJET | Impact Factor value: 4.45 | ISO 9001:2008 Certified Journal | Page 36 Survey on Artificial Neural Network Learning Technique Algorithms K Ishthaq Ahamed 1 , Dr. Shaheda Akthar 2 1 Faculty, Department of CSE, G. Pulla Reddy Engineering College, Kurnool, AP, India. 2 Faculty of Computer Science, Govt. College for Women, Guntur, AP, India. ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - Automation is now ubiquitous, and research for developing new ones is constantly growing. It plays an important role in decision-making, prediction, classification to provide a fast and accurate result. Due to their adaptive learning and nonlinear mapping properties, the artificial neural networks are widely used to support the human decision capabilities, avoiding inconsistency in practice and errors based on lack of experience. Neural networks is a mathematical model of neurons in human brain, possess all abilities mention above, also they provide parallel computations. Learning rules are algorithms which direct changes in the weights of the connections in a network. They are incorporating an error reduction procedure by employing the difference between the desired output and an actual output to change its weights during training. The learning rule is typically applied repeatedly to the same set of training inputs across a large number of epochs with error gradually reduced across epochs as the weights are fine-tuned. In this paper error correction, memory based, hebbian and competition learning rules are explored for better predictions and learning. Key Words: Artificial neural networks, error correction, memory based, hebbian, and competition learning. 1. Introduction Models has constructed by computational neurobiologists by using neurons to simulate the behavior of the brain. As Computer Scientists, we are more interested in the general properties of neural networks, independent of how they are actually "implemented" in the brain. This means that we can use much simpler, abstract "neurons", which capture the essence of neural computation even if they leave out much of the details of how biological neurons work. People have implemented model neurons in hardware as electronic circuits, often integrated on VLSI chips. Remember though that computers run much faster than brains, therefore we can run fairly large networks of simple model neurons as software simulations in reasonable time [10]. This has obvious disadvantages over having to use special "neural" computer hardware. The basic of neuron model is often known as node or unit. It receives input from some other units, or perhaps from an external source. Each input has an associated weight w, which can be modified so as to model synaptic learning. The unit computes some function f of the weighted sum of its inputs. Its output, in turn, can serve as input to other units. The weighted sum is called the net input to unit i, often written neti. Weight from unit j to unit i is denoted as wij. The function f is the unit's activation function. In the simplest case, Fig.1, f is the identity function, and the unit's output is just its net input. This is called a linear unit. Fig 1: Simple Artificial Neuron Model Neural Network Applications can be grouped in following categories: Prediction system The task is to forecast some future values of a time- sequenced data. Prediction has a significant impact on decision support systems. Prediction differs from function approximation by considering time factor [1][2]. Clustering A clustering algorithm explores the relationship between patterns and places related patterns in a cluster. Well known applications are data compression and data mining [3]. Classification recognition Classification is to assign an input pattern (like handwritten symbol) to one of many classes. This category consists of algorithmic implementations like associative memory [5]. Function approximation The main use of function approximation is to estimate of the unknown function f( ) subject to noise. In engineering various methods require function approximation. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 03 Issue: 02 | Feb-2016 www.irjet.net p-ISSN: 2395-0072